north of dumbarton bridge copper and nickel site-specific ... · north of dumbarton bridge copper...

D E C E M B E R 2 0 0 4

Clean Estuary Partnership

North of Dumbarton BridgeCopper and Nickel Site-SpecificObjective (SSO) Derivation

Prepared by:

EOA, Inc.

LARRY WALKER ASSOCIATES

L A R R Y

W A L K E R

ASSOCIATES

Site-Specific Objective Derivation Report March 2005i

Table of Contents

Glossary of Acronyms........................................................................................... ivExecutive Summary ............................................................................................... v

1. Introduction........................................................................................................ 11.1 USEPA SSO Calculation Methodologies...........................................................................1

1.1.1 Recalculation Procedure .............................................................................................11.1.2 Indicator Species Procedure........................................................................................11.1.3 Resident Species Procedure ........................................................................................2

2. San Jose Nickel SSO Approach .......................................................................... 2

3. Copper Site-Specific Objective Development and Selection ............................... 43.1 Laboratory Procedures ......................................................................................................43.2 WER Results.....................................................................................................................6

3.2.1 Individual WER Results .............................................................................................63.2.2 Pooled WER Results ................................................................................................ 113.2.3 Additional WER Data Statistical Analyses ............................................................... 123.2.4 Site-to-Site Variations .............................................................................................. 143.2.5 Sample Specific WER Results.................................................................................. 16

3.3 San Jose Approach to Final WER and SSO Selection...................................................... 173.3.1 FWER Selection....................................................................................................... 173.3.2 Recalculation of the National Copper Criterion ........................................................ 173.3.3 San Jose SSO Selection ............................................................................................ 18

3.4 NDB Final WER and SSO Selection ............................................................................... 193.4.1 Comparison of North Of Dumbarton Study and San Jose Study WER Results.......... 203.4.2 NDB Site-Specific CCC (SSO) Recommendation .................................................... 213.4.3 Revised SSO Recommendation ................................................................................ 22

4. Biotic Ligand Model......................................................................................... 23

5. Compliance Evaluation with SSO Based Effluent Limits.................................. 26

6. References........................................................................................................ 29

Site-Specific Objective Derivation Report March 2005ii

Appendices

Appendix A. Individual Sample Station Calculated SSOs Compared to Ambient DissolvedCopper Concentrations

Appendix B. Executive Summary to Bay Area Clean Water Agencies Copper & Nickel North ofthe Dumbarton Bridge Step 1: Impairment Assessment Report AmbientConcentrations and WERs September 2000 – June 2001

Appendix C. Response to CommentsAppendix D. June 3, 2004 Copper & Nickel Workgroup Meeting Materials

- June 3, 2004 Copper & Nickel Workgroup Meeting Agenda- June 3, 2004 Copper & Nickel Workgroup Meeting Notes- San Jose PowerPoint Presentation from June 3, 2004 Copper & Nickel Workgroup Meeting:

Selection of NDB Copper WERs: Use Of The Mytilus Embryo Assays to Derive SSOs for SanFrancisco Bay North of Dumbarton Bridge

- San Jose PowerPoint Presentation from June 3, 2004 Copper & Nickel Workgroup Meeting:Development of a S.F. Bay Site-Specific Chronic Criterion for Nickel Using the EPARecalculation Procedure and Modification of the EPA Nickel Saltwater Acute-To-ChronicRatio

Site-Specific Objective Derivation Report March 2005iii

List of Tables

Table 1. Dissolved Copper WER Summary Statistics by Event..................................................8Table 2. Individual Station Dissolved Copper WER Median Values ..........................................8Table 3. Dissolved Copper Pooled WER Summary Statistics ................................................... 12Table 4. Mean WERs with 95% Confidence Intervals.............................................................. 13Table 5. Major Subsections of San Francisco Bay North of the Dumbarton Bridge .................. 14Table 6. Subsection Comparisons and P-values........................................................................ 15Table 7. Sample-Specific WER Approach Results ................................................................... 16Table 8. Current and Modified National Copper Criteria to reflect the addition of NDB WERstudy data into the San Jose study modified EPA National Copper Criteria database................. 19Table 9. Comparison of North of Dumbarton and San Jose Dissolved Copper WER Results..... 20Table 10. NDB Pooled Station Dissolved Copper SSO Summary Statistics............................... 22Table 11. Copper SSO Based Effluent Limits – Case Study Compliance Evaluation................. 27Table 12. Nickel SSO Based Effluent Limits – Case Study Compliance Evaluation .................. 27Table 13. Copper SSO Based Effluent Limits – All Dischargers Compliance Evaluation Shallow Water Dischargers Compliance.................................................................... 28Table 14. Nickel SSO Based Effluent Limits – All Dischargers Compliance Evaluation ........... 28

List of Figures

Figure 1. Dissolved Copper WER at Each Station for Each Event (µg/L)....................................9Figure 2a. Dissolved Copper WER Results by Sampling Station............................................... 10Figure 2b. Dissolved Copper WER Results by Event ................................................................ 10Figure 3. Mean WERs: Deep versus Shallow............................................................................ 13Figure 4. Mean WERs: Deep versus Shallow for Each Event .................................................... 14Figure 5. Mean WERs: Event (a.) and Overall (b.).................................................................... 14Figure 6. Mean WERs in Distinct Areas of San Francisco Bay ................................................. 15Figure 7. RMP Newly Defined Regions of San Francisco Bay .................................................. 23Figure 8. Comparison of BLM Predicted versus NDB Study Measured Toxicity(Mytilus EC50) ......................................................................................................................... 25

Site-Specific Objective Derivation Report March 2005iv

GLOSSARY OF ACRONYMSµg/L Micrograms per liter, or parts per billionACR Acute-to-Chronic Ratio

AMEL Average Monthly Effluent LimitANOVA Analysis of VarianceBACWA Bay Area Clean Water Agencies

BASMAA Bay Area Stormwater Management Agencies AssociationBLM Biotic Ligand Model

CC Coordinating CommitteeCCC Criterion Continuous ConcentrationCMC Criterion Maximum Concentration

CMIA Conceptual Model and Impairment Assessment ReportCTR California Toxics Rule

Cu CopperCV Coefficient of variation

DOC Dissolved Organic CarbonEBMUD East Bay Municipal Utility District

EC50 50% Effect Concentration (concentration that adversely effects 50% of the species tested)FACR Final Acute-to-Chronic Ratio

FAV Final Acute ValueFSSD Fairfield Suisun Sanitary District

FWER Final Water-Effects RatioLGVSD Las Gallinas Valley Sanitary District

LSB Lower South BayMEC Maximum Effluent ConcentrationNDB North of Dumbarton Bridge

Ni NickelPER Pacific EcoRisk Laboratory

POTW Publicly Owned Treatment Worksppb Parts per billionppt Parts per thousand

p-value Significance probability valueQA/QC Quality Assurance/Quality Control

RMP Regional Monitoring ProgramRWQCB Regional Water Quality Control Board

SAIC Science Applications International CorporationSDB South of the Dumbarton BridgeSIP Policy for Implementation of Toxics Standards for Inland Surface Waters, Enclosed Bays,

and Estuaries of California; aka State Implementation PolicySMAV Species Mean Acute Value

SSO Site-Specific ObjectiveTMDL Total Maximum Daily Load

TOC Total Organic CarbonTRC Technical Review CommitteeTSS Total Suspended Solids

U.S. EPA United States Environmental Protection AgencyWER Water-Effects RatioWQC Water Quality CriteriaWQO Water Quality Objective

WSPA Western States Petroleum Agencies

Site-Specific Objective Derivation Report March 2005v

EXECUTIVE SUMMARY

IntroductionThis report describes the methodologies and rationale for the establishment of site-specific waterquality objectives for copper and nickel in San Francisco Bay North of the Dumbarton Bridge(NDB). Methodologies used conform with U.S. EPA guidance for development of site-specificobjectives and are consistent with approaches used in the development and approval of site-specific objectives for copper and nickel in the Lower South Bay.

USEPA SSO Calculation MethodologiesBecause a national aquatic life criterion might be more or less protective than intended for theaquatic life in most bodies of water, the U.S. EPA has provided guidance concerning threeprocedures that may be used to derive a site-specific criterion (U.S. EPA, 1994). Theseprocedures are discussed in this report

San Jose Nickel SSO ApproachFor nickel, a combination of the Recalculation procedure and modification of the U.S. EPArecommended Acute-to-Chronic Ratio (ACR) was used by San Jose to develop site-specificmodifications to the national water quality criterion. In 1995, Watson, et al. (1996) recalculatedthe numeric nickel national water quality criterion using the procedure outlined by the U.S. EPA(Carlson, et al. 1984). The corrections, additions, and deletions resulted in a proposed criterion of10.2 µg/L using the most conservative approach. During this recalculation process, it becameobvious that there were no recent chronic data that could be used to recalculate the Final Acute-to-Chronic Ratio (FACR).

In 1997, Watson, et al. (1999) designed and conducted acute and chronic flow-through bioassaytests on three marine species (topsmelt fish, Atherinops affinis; red abalone, Haliotes rufescens;and the mysid shrimp, Mysidopsis intii). The resultant acute-to-chronic ratios for all three marinespecies tested by San Jose were remarkably similar, ranging from 5.50 to 6.73. These valueswere in turn comparable to the ACR value previously reported for M. bahia of 5.48 (U.S. EPA1986). A FACR derived solely from a geometric mean of these four marine species ACRs wouldbe 5.959. An alternative FACR of 10.50 was also developed, using a combination of the fourmarine ACRs plus two freshwater ACRs.

Watson, et al (1996, 1999) updated the national data-set by deleting non-native species,eliminating questionable data from the data set, adding additional saltwater acute and chronic testdata to the dataset, and recalculating both new “proposed” national and site-specific criteria fornickel.

Copper Site-Specific Objective Development and SelectionThe Copper Site-Specific Objectives (SSO) Development and Selection Section presents a briefsummary of the copper Water Effect Ratio (WER) chemical and toxicological methods used inthe North of Dumbarton Bridge (NDB) WER study, followed by a presentation of the individualstation and sampling event WER results, the pooled station WER results, and then use of theWER results to derive a range of potential SSOs for the Bay NDB.

Site-Specific Objective Derivation Report March 2005vi

WER ResultsThe NDB WER study developed and presented the WER results by individual station and event(paired bar graphs); pooled by station (all four events); pooled by event (all 13 stations); andpooled for All sites, North Bay and Central Bay. In addition, a more limited analyses for All sites(except BD15); and shallow water and deep water sites (given their lower significance as agrouping factor) was developed and presented.

San Jose Approach to Final WER and SSO SelectionThis section reviews the approach and reasoning used by the City of San Jose in their WERstudy (May 1998) in evaluating final WERs (FWERs) and calculating alternative SSOs for thebay SDB. The discussion below focuses on the reasoning employed when evaluating the prosand cons of different data (station) pooling alternatives. Many of the variables assessed by SanJose for SDB are relevant to the FWER decisions to be made NDB.

NDB Final WER and SSO SelectionThe current copper WQO applicable to San Francisco Bay NDB is the CTR CCC value of 3.1µg/L times a WER (the default WER is 1.0). Appendix A shows the complete set of individualNDB calculated site-specific WER based SSOs for each of the four events at each samplingstation. These are the copper objective alternatives that are directly sanctioned by the CTR.

The CTR WQO based SSOs for all four events ranged from 5.2 µg/L (BF20) to 8.4 µg/L (BA40and BD15). While some of the ambient copper values approached the non-WER adjusted 3.1µg/L level, most would be a factor of two to three below a SSO based on the WERs developed inthe NDB study and either the CTR or recalculated WQOs.

NDB Site-Specific CCC (SSO) RecommendationA primary goal of the NDB WER study was to produce scientifically defensible WER valuesthat could be used with confidence by State and U.S. EPA regulators, dischargers andstakeholders to establish one or more SSOs for the Bay north of Dumbarton Bridge. Severalconservative measures were employed in both studies including: using M. edulis, the mostsensitive species listed in the marine criteria data set for copper, as the test species; andconsideration of lowering the national CCC from 3.1 to 2.5 µg/L dissolved copper byincorporating the site-specific laboratory water results into the national copper data set.

The U.S. EPA guidance suggests using geometric means for FWER selection. The arithmeticmeans ranged from 2.5 to 2.8 while the geometric means ranged from 2.4 to 2.7. The All Sitesarithmetic and geometric means are 2.7 and 2.6, respectively, in the middle of the alreadyrelatively narrow Central to North Bay range cited. The prior statistical analysis had shown thereto be no significant differences between results at shallow versus deep water stations so thosegroupings are not considered further in the SSO selection analysis.

The prior statistical analysis found only a minor difference in WERs (0.5) between a pooling ofLower Bay versus San Pablo Bay stations. It further found that 0.5 was approximately thedifference between the upper and lower 95% confidence intervals for the All Sites pooled WERalternative. Additional analysis showed there to be no statistically significant difference betweenthe Central Bay and the North Bay pooled WERs.

Site-Specific Objective Derivation Report March 2005vii

These relatively small differences between the various pooled dataset provides some support forselecting a single NDB SSO using all the available data. If arithmetic averages are used (giventhat the data are normally distributed), an All Sites NDB SSO could range from 6.8 to 8.4 µg/L,depending on whether the CTR or recalculated WQO is used as the basis of adjustment by theAll Sites FWER of 2.7.

Revised SSO RecommendationSince the time of the WER study, the RMP has reevaluated the regional definitions in the Bay.The RMP now recognizes 5 regions:

1. Suisun Bay2. San Pablo Bay3. Central Bay4. South Bay5. Lower South Bay

Rather than keeping the SDB work separate from the NDB work, it has been found that it ismore appropriate to integrate the studies and create two SSOs for the entire Bay. These potentialSSOs are calculated as 6.0 ppb for Bay Regions 1-3 and 6.9 ppb for Bay Regions 4 & 5. Thisapproach protects Mytilus sp., the most sensitive species in the U.S. EPA database and acommercially important species. These SSOs are the result of two proposed WER values (2.4 forRegions 1-3; 2.7 for Regions 4-5).

Biotic Ligand ModelThe Biotic Ligand Model (BLM) predicts metal toxicity to aquatic organisms based on thechemical characterization of a given water body. The model takes into consideration severalwater quality parameters, including hardness, DOC, chloride, pH, and alkalinity. The BLM wasused to predict copper toxicity in the NDB WER study water samples from San Francisco Bay,and in the laboratory water samples from the Granite Canyon Marine Laboratory in Carmel, CA.BLM input chemistry was measured in the second, third, and fourth sampling events. This modelwas previously developed with toxicity data from South San Francisco Bay WER study (Paquinet al., 2000). The model was used to predict EC50s “blind”, i.e. without knowing the measuredvalues until after predictions were made.

Compliance Evaluation with SSO Based Effluent LimitsThe SIP SSO Justification Request Report (September 2004) included an evaluation of the abilityof three case study POTWs to comply with non-WER adjusted CTR based copper and nickeleffluent limits. Copper compliance continues to be an issue for shallow water (zero dilution)municipal secondary treatment plants such as LGVSD no matter what WER/SSO is selected.Copper compliance may also continue to be an issue also for shallow water advanced secondaryplants such as FSSD, depending on the SSO selected. Deepwater secondary treatmentdischargers (with 10:1 dilution) with performance equivalent to EBMUD would appear to haveminimal compliance issues with any SSO based limit.

Site-Specific Objective Derivation Report March 2005viii

Secondary treatment POTWs and industries without dilution credit will have moderate tosignificant copper compliance problems even with the upper range of SSO based effluent limits.Advanced secondary POTWs without dilution may have minor compliance problems if relativelylow WER based SSOs are selected. A small percentage of facilities with 10:1 dilution may havecopper compliance problems if relatively low copper WER based SSOs are selected.

Site-Specific Objective Derivation Report March 20051

1. INTRODUCTIONThis report describes the methodologies and rationale for the establishment of site-specific waterquality objectives for copper and nickel in San Francisco Bay North of the Dumbarton Bridge(NDB). Methodologies used conform with U.S. EPA guidance for development of site-specificobjectives and are consistent with approaches used in the development and approval of site-specific objectives for copper and nickel in the Lower South Bay. Information used to developsite-specific objectives NDB was developed through a peer reviewed effort that was coordinatedwith interested parties including the Regional Water Quality Control Board, U.S. EPA RegionIX, Bay Area Clean Water Agencies (BACWA), Bay Area Stormwater Management AgenciesAssociation (BASMAA), Western States Petroleum Association (WSPA), San FranciscoBaykeeper, and the Copper Development Association.

1.1 USEPA SSO Calculation Methodologies

Because a national aquatic life criterion might be more or less protective than intended for theaquatic life in most bodies of water, EPA has provided guidance concerning three proceduresthat may be used to derive a site-specific criterion (U.S. EPA, 1994):

1.1.1 Recalculation Procedure

The Recalculation Procedure is intended to take into account relevant differences between thesensitivities of the aquatic organisms in the national dataset and the sensitivities of organismsthat occur at the site. This procedure involves eliminating non-resident species from the nationaldata set of aquatic species whose toxicity test results are used to compute the water qualitycriterion, and then recalculating a site-specific objective with the modified set of species.

1.1.2 Indicator Species Procedure

The Indicator Species procedure is based on the assumption that characteristics of ambient watermay influence the bioavailability and toxicity of a pollutant. Acute toxicity in site water andlaboratory water is determined in side-by-side toxicity tests using either resident species oracceptable sensitive non-resident species, which are used as surrogates for the resident species.The Indicator Species Procedure allows for modification of the national criterion by using a site-specific multiplier that accounts for ambient water quality characteristics that may affect thebioavailability of the pollutant in question. As part of this procedure, a water effects ratio (WER)is determined using results from toxicity tests performed in ambient water and laboratory water.

A WER is the ratio of toxicity of a compound to an aquatic organism when the tests areperformed using standard laboratory water versus the toxicity when the tests are performed usingambient water. A WER is expected to appropriately take into account the (a) site-specifictoxicity of a compound and (b) interactions with other constituents of the site water that mayeither reduce or increase the toxicity of the compound in question. If the value of the water effectratio exceeds 1.0, the pollutant is less toxic in the site water than in laboratory water. Thedifference in toxicity values, expressed as a WER, is used to convert a national water qualitycriterion for a pollutant to a site-specific water quality criterion.


The City of San Jose used the Indicator Species Procedure in its Impairment Assessment forcopper. Observed WER values ranged from 2.5 to 5.2 based on measured dissolved copper. Therecommended range of chronic SSOs for the lower South Bay resulting from the ImpairmentAssessment was 5 to 12 µg/L dissolved copper. U.S. EPA reviewed this work and found that thespecies used were appropriate, the data valid and the conclusions reasonable (USEPA July 27,1998).

1.1.3 Resident Species Procedure

This procedure is used to account for differences in resident species’ sensitivity and differencesin bioavailability and toxicity of a material due to the physical and chemical characteristics of theambient water. The Resident Species Procedure allows for modification of the national criterionby concurrently testing resident species for chronic and acute toxicity in ambient site water.

2. SAN JOSE NICKEL SSO APPROACHFor nickel, a combination of the Recalculation procedure and modification of the U.S. EPArecommended Acute-to-Chronic Ratio (ACR) was used by San Jose to develop site-specificmodifications to the national water quality criterion. In 1995, Watson, et al. (1996) recalculatedthe numeric nickel national water quality criterion using the procedure outlined by the U.S. EPA(Carlson, et al. 1984). The corrections, additions, and deletions resulted in a proposed criterion of10.2 µg/L using the most conservative approach. During this recalculation process, it becameobvious that there were no recent chronic data that could be used to recalculate the Final Acute-to-Chronic Ratio (FACR).

The FACR derived in 1986 (17.99) was based on two freshwater and one marine species. Therewas a large difference between the freshwater and saltwater ACR values that contributed to theFACR. The ACR for the freshwater minnow, Pimephales promelas, was 35.58 and that for thewaterflea, Daphnia magna, was 29.86. Only one marine species, the mysid shrimp, Mysidopsisbahia (since reclassified as Americamysis bahia), had verifiable chronic data which resulted in asingle marine ACR value of 5.48.

In 1997, Watson, et al. (1999) designed and conducted acute and chronic flow-through bioassaytests on three marine species (topsmelt fish, Atherinops affinis; red abalone, Haliotes rufescens;and the mysid shrimp, Mysidopsis intii). The topsmelt is a native to Lower South San FranciscoBay, while the other two species are West Coast natives and commonly used surrogate residentspecies. Abalone and mysids were found to be far more sensitive to nickel than was topsmelt.Chronic values for abalone and mysids were similar (26.43 and 22.09 µg/L, respectively), andwere lower than available literature values. The chronic value for the topsmelt was 4,270 µg/L.

The resultant acute-to-chronic ratios for all three marine species tested by San Jose wereremarkably similar, ranging from 5.50 to 6.73. These values were in turn comparable to the ACRvalue previously reported for M. bahia of 5.48 (U.S. EPA, 1986). A FACR derived solely froma geometric mean of these four marine species ACRs would be 5.959. An alternative FACR of10.50 was also developed, using a combination of the four marine ACRs plus two freshwaterACRs.


Watson, et al (1996, 1999) updated the national data-set by deleting non-native species,eliminating questionable data from the data set, adding additional saltwater acute and chronic testdata to the dataset, and recalculating both new “proposed” national and site-specific criteria fornickel.

Since abalone is a commercially important species, the calculated Final Acute Value (FAV) thatwould normally be used for criteria derivation was replaced in the national dataset by the lower(more conservative) abalone Species Mean Acute Value (145.5 µg/L) in order to protect thisspecies. Thus, the recalculated potential national and “South San Francisco Bay” site-specificFAVs were 145.5 µg/L and 124.8 µg/L, respectively. While the San Jose reports used theterminology “South San Francisco Bay” SSOs, the approach taken resulted in a range of SSOvalues applicable throughout the Bay and potentially to the West Coast. This report will use the“Resident Species” terminology for this SSO approach.

Using the two updated FACRs (marine and combined freshwater plus marine) and the tworecalculated FAVs (national and resident species), four alternative SSOs can be derived using theFormula: FAV ÷ ACR = CCC

1) Recalculated National Criterion/Combined Freshwater and Marine ACR; 145.5 µg/L ÷ 10.50 = 13.86 µg/L

2) Recalculated National Criterion/Marine ACR145.5 µg/L ÷ 5.959 = 24.42 µg/L

3) SF Bay Resident Species/Combined Freshwater and Marine ACR; and124.8 µg/L ÷ 10.50 = 11.89 µg/L

4) SF Bay Resident Species/Marine ACR;124.8 µg/L ÷ 5.959 = 20.94 µg/L

The chronic values of 22.09 and 26.43 ug /L for mysids and abalone, respectively indicate thatall but option 2) (24.42 µg/L) of the above four potential nickel SSOs would be protective (inclean laboratory water) of the more sensitive mysid (and abalone) and, as such, be protective ofthe Beneficial Uses San Francisco Bay and North and South of the Dumbarton Bridge. It shouldbe noted, however, that these SSO values are based on clean laboratory toxicity test results anddo not include any of the ambient “apparent complexing capacity” present in the Bay that may beresponsible for making nickel even less bioavailable to aquatic organisms.

The U.S. EPA reviewed this San Jose work and found that the species and methodologies usedwere appropriate for developing site-specific modifications to the national water quality criterionfor nickel. As such, no additional toxicity testing is required to derive a nickel SSO for otherregions of the Bay. Use of the resident species dataset, while more conservative, would appearappropriate for establishing a NDB SSO, versus use of the recalculated national dataset.


Decisions are required as to whether it is more technically appropriate to use the four speciesmarine ACR versus the combined freshwater/marine (used for the LSB) given the relativerobustness of the marine ACR dataset.

3. COPPER SITE-SPECIFIC OBJECTIVE DEVELOPMENT ANDSELECTIONThis Copper Site-Specific Objectives (SSO) Development and Selection Section presents a briefsummary of the copper Water Effect Ratio (WER) chemical and toxicological methods used inthe North of Dumbarton Bridge (NDB) WER study, followed by a presentation of the individualstation and sampling event WER results, the pooled station WER results, and then use of theWER results to derive a range of potential SSOs for the Bay NDB.

3.1 Laboratory Procedures

To address the aquatic toxicity of copper and nickel in San Francisco Bay, well-definedsampling, laboratory and quality assurance/quality control (QA/QC) procedures were used in theNDB Water Effects Ratio (WER) Study, based in large part on the San Jose WER studies.Detailed descriptions and information relating to sampling, laboratory and QA/QC proceduresare provided in Sections 2 through 4 (and associated Appendices 2 through 4) of the NDB copperWER study (July 2002) and in Appendix 1 (Study Work Plan) of that study. The procedures andresults of NDB WER study were reviewed by the Coordinating Committee (CC) and theTechnical Review Committee (TRC) as documented in the CC meeting notes (provided withinAppendix 1of the NDB WER study) and the response to the TRC comments on the interim andthe draft final report (Appendix 8 of the NDB WER study).

The NDB copper WER study closely followed the basic San Jose WER study approach by usingthe indicator species Mytilus edulis as the test organism. The Mytilus edulis toxicity test used forthe North of Dumbarton Bridge WER study (NDB WER study) followed the guidelinesestablished by the USEPA manual [U.S. EPA, 1995b]. M. edulis is an almost ideal organism foruse in WER copper studies. When deriving a site-specific criterion, it is desirable to use a testspecies that is sensitive at Criterion Continuous Concentrations (CCC) or Criterion MaximumConcentrations (CMC). The concentrations that affected M. edulis approximate the criteriaconcentrations. M. edulis is the most appropriate species to use as a surrogate for brackish waterspecies that inhabit the North Bay and for setting a North Bay site-specific criterion for copper.This conclusion is based on several factors:

• The CTR criterion for copper is determined exclusively by M. edulis for protection of acommercially important species. Since it is used exclusively for setting the current nationalcriterion, it is appropriate to use it exclusively for setting a site-specific criterion for theNorth Bay.

• It is the most sensitive species in the national saltwater database. It is not only a goodsurrogate for invertebrate species (which tend to be more sensitive to copper thanvertebrates) and mollusks (a phylum sensitive to copper – the 3rd, 4th, and 6th most sensitivespecies in the national copper database are mollusks), but it is a good surrogate for anysensitive saltwater animal (at any salinity above ~ 2 parts per thousand (ppt)).


• The most sensitive freshwater species to copper are daphnids (water fleas). In soft water(where copper is more bioavailable), they are about as sensitive as M. edulis (Genus MeanAcute Value (GMAV) of 14.48 parts per billion (ppb) for the genus Daphnia, 9.92 ppb forCeriodaphnia and 9.63 ppb for Mytilus). However, daphnids would be poor surrogates foranimals living in brackish water (e.g., at typical 5 ppt salinity at BF 10 and BF 20 sites) sincethe acute toxicity values for freshwater animals are more significantly dependent on hardnessthan saltwater animals. For example, the estimated acute value for Ceriodaphnia (at ahardness of 5 ppt salinity seawater) would be so high as to be effectively meaningless.

The methodology for copper spiking and test solution preparation was developed in conjunctionwith San Jose researchers. Water used for laboratory water and reference toxicant tests was 1 µmsand-filtered natural seawater obtained from the Granite Canyon Marine Laboratory in Carmel,CA. Test concentrations were prepared by spiking one-liter aliquots of the salinity-adjustedlaboratory and site waters with a certified commercial copper nitrate standard. To confirm thatMytilus edulis embryos were responding to toxic stress in a typical fashion, a reference toxicanttest was run concurrently with each set of site water (and lab water) tests. All reference toxicantresults were within acceptable limits (±2 standard deviations about the mean).

Once toxicity testing was completed, guidance within the U.S. EPA memorandum entitledInterim Guidance on the Determination and Use of Water Effect Ratios for Metals was used toselect test solutions for chemical analysis [U.S. EPA, 1994]. Consistent with the City of SanJose’s study, WER calculations were based on initial copper concentrations as opposed to a time-weighted average of initial and final values. San Jose studies (and the TRC) found this approachto be more conservative since a proportionately greater copper recovery is expected in site waterthan in lab water when measured at the test conclusion [San Jose, 1998].

EC50 values were calculated using the Trimmed Spearman-Karber Method. EC50 values fortotal and dissolved copper in lab water exhibited high precision, with a coefficient of variation(CV) of 16.1% and 17.3% respectively. This compares favorably with the CV of 23.1% and22.0%, respectively reported for the City of San Jose study.

Dissolved copper EC50 values were used to calculate the WERs for each station and event:

There were a total of 50 valid site water EC50s and eight lab water EC50s developed in the NDBWER study. There were two laboratory water results developed for each event, to coincide withthe Central Bay and North Bay samples were collected and run on separate days.

The NDB WER study and associated analyses were performed consistent with RMP-typemonitoring and analysis activities, some of which use research based methods to obtain thehighest quality data possible. Rigorous quality control/quality assurance practices weremaintained during all aspects (sampling, testing, chemical analysis) of the NDB WER study.

WER =

Site Water EC50

Lab Water EC50


This is evidenced by the high quality, low variability results obtained in compliance with theindividual lab’s QA/QC criteria.

The eight laboratory water tests all generated results acceptable for calculating WERs and forcalculating national criteria. The laboratory water EC50 value (Final Acute Value (FAV)) usedto derive the national WQC for copper is 9.625 µg/L. This FAV is based on the M. edulis SMAVof 9.625. The WER guidance document (U.S. EPA 1994) defines laboratory water test results asbeing acceptable if they are within a factor of 1.5 of the national results (i.e., 6.417 to 14.468 forcopper). The eight PER laboratory water results readily met this criterion since they are within afactor of about 1.25 of the national results (PER arithmetic and geometric means of 7.75 and7.66, respectively). In addition, the PER lab water results were quite consistent with a 1.28standard deviation and 16.5% coefficient of variation.

Overall, the results from the four sampling events were found to have sufficient QA to supportthe reported chemical and toxicological bioassay data. The Coordinating Committee andTechnical Review Committee reviewed the WER report Work Plan and methods; the resultsafter the first sampling event (per WER guidance); and the NDB WER study final results. TheTRC comments and the project team’s response to comments are summarized in the NDB WERstudy and included in their entirety within Appendix 8 of the NDB WER study. The TRC foundthat the WER and associated data were of high quality and suitable to be used for calculatingsite-specific objectives.

3.2 WER Results

The NDB WER study developed and presented the WER results by individual station and event(paired bar graphs); pooled by station (all four events); pooled by event (all 13 stations); andpooled for All sites, North Bay and Central Bay. In addition, a more limited analyses for All sites(except BD15); and shallow water and deep water sites (given their lower significance as agrouping factor) was developed and presented.

3.2.1 Individual WER Results

The NDB WER study developed 50 overall WERs (four events, thirteen stations per event witheleven valid results in Event 4). Dissolved copper WERs at each station (Figure 1) showedgeneral consistency between events, except at stations BD15, LCB01, LCB02, and BA40 wherethere were moderate to significant spikes during Event 2. In addition, total suspended solids(TSS), total organic carbon (TOC), and dissolved organic carbon (DOC) concentrations trendedhigher at these sites during Event 2. Spatially there were no readily discernible trends or patterns.The Grizzly Bay station (BF20) had some of the lowest WERs while the next closest stationPacheco Creek (BF10) had some moderately high values. The mouth of the Petaluma River(BD15) station had consistently elevated values (see discussion in CMIA report). The othernorthern and central bay station WERs were typically in the 2-3 range with the exception of thesouthern most station Redwood Creek (BA40) with values closer to 3.

Box and whisker plots have been used for the various pooled data presentations to show themedian, the 25th percentile, the 75th percentile, extreme values and outliers. The lower and upperboundaries of the box represent the 25th and 75th percentiles, respectively. The horizontal line


inside the box represents the median. The length of the box corresponds to the inter-quartilerange, which is the difference between the 75th and 25th percentiles. The box plot includes twocategories of cases with outlying values. Cases with values that are more than three box-lengthsfrom the upper or lower edge of the box are designated extreme values and are shown withasterisks. Cases with values that are between 1.5 and 3 box-lengths from the upper or lower edgeof the box are outliers and shown with circles. The largest and smallest observed values that arenot outliers are also shown. Lines (referred to as whiskers) are drawn from the ends of the box tothese values.

The upper plot (Figure 2a) shows spatial (station-by-station) results and the lower plot (Figure2b) shows temporal (event-by-event) results. It is important to note that the station-by-stationboxes include only four data points. Therefore, the 25th and 75th percentile values have minimalsignificance in these plots. However, these are still useful for illustrating differences betweenstations.

Figure 2a shows that there is some variability in dissolved Cu WERs from site-to-site but againno clear-cut spatial pattern or trend. The median WER values range only between approximately2 and 3, while the smallest and largest observed values range from approximately 1.5 to 5.5.BD15 showed the most variability in dissolved Cu WER. BB15 and BC10 showed only slightvariation from event-to-event. Based on the position of the median bar, the BA40, BB15,LCB01, LCB02, BB30, BD15 and BF20 WER data are skewed slightly negatively. That is, thedark line in the box is not in the center but closer to the bottom of the box. This is reflective ofthe significant difference in Event 2 WERs compared to the consistent WERs during the otherthree events at these stations. For example, at LCB01 the event 2 WER was 4.7 versus 2.5, 2.4,and 2.4 during other three events (Figure 1 and Table 1).

The individual station dissolved copper WER results are pooled and summarized by event inFigure 2b. The main conclusion to be drawn is that when compared to the other three events,Event 2 had higher dissolved Cu WER values at most stations. Median values for all sitescombined for each of the four events ranged from approximately 2.5 to 3.2. Data are slightlyskewed (positive) for Events 1 and 3. Events 1, 3 and 4 do not show major variability based onthe comparatively short lengths of the boxes and whiskers. However, Event 2 showed dissolvedCu WER values ranging from approximately 2.5 to 5.5.

General summary statistics were calculated for the individual stations and events and reviewedfor evidence of patterns or trends (Table 1). The two highest WERs recorded during this studyoccurred during Event 2 at the mouth of the Petaluma River (BD15) (5.3) and at Pacheco Creek(BF10) (3.1). The two lowest occurred at San Pablo Bay (BD20) (1.5) during Event 4 and atGrizzly Bay (BF20) (1.6) during Event 3. Overall median WERs by event were 2.4, 3.2, 2.7, and2.2 for Events 1 through 4, respectively. The overall grand median WER was 2.5. Stationdissolved copper WER median values are presented in Table 2. Median values ranged from 1.7at BF20 to 3.1 at the adjacent BF10 station.


Table 1. Dissolved Copper WER Summary Statistics by Event

Station Event 1 Event 2 Event 3 Event 4BA40 2.7 4.2 2.7 3.1BB15 2.4 3.2 2.7 2.5

LCB01 2.5 4.7 2.4 2.4LCB02 2.4 5.2 2.8 2.2BB30 2.5 3.5 2.4 2.4C

entr

al B

ayBC10 2.2 2.6 2.4 1.8BD20 2.2 2.6 2.0 1.5SPB01 2.0 2.6 2.9 2.0BD15 2.7 5.3 3.4 2.4SPB02 1.7 3.2 2.4 2.2SPB03 1.7 2.5 2.7 2.1BF10 2.5 3.5 3.1 *N

orth

Bay

BF20 1.7 3.2 1.6 *number 13 13 13 11minimum 1.7 2.5 1.6 1.5maximum 2.7 5.3 3.4 3.1a. mean 2.3 3.5 2.6 2.3g. mean 2.2 3.4 2.5 2.290th Percentile 2.7 5.1 3.1 2.55th Percentile 1.7 2.5 1.8 1.7median 2.4 3.2 2.7 2.2Std. deviation 0.4 1.0 0.5 0.4

*data did not meet QA/QC criteria and were not used in calculations

Table 2. Individual Station Dissolved Copper WER Median Values

Station MedianBA40 2.9BB15 2.6

LCB01 2.5LCB02 2.6BB30 2.5C

entr

al B

ay

BC10 2.3BD20 2.1SPB01 2.3BD15 3.0SPB02 2.3SPB03 2.3BF10 3.1N

orth

Bay

BF20 1.7


Figure 1. Dissolved Copper WER at Each Station for Each Event (µg/L)

0

1

2

3

4

5

6

BA40 BB15 LCB01 LCB02 BB30 BC10

WER

Event 1Event 2Event 3Event 4

Redwood Creek San Bruno L. Central Bay (midpoint)

L. Central Bay (nearshore) Oyster Point Yerba Buena Island

0

1

2

3

4

5

6

BD20 SPB01 BD15 SPB02 SPB03 BF10 BF20

WER

Event 1Event 2Event 3Event 4

San Pablo Bay Between BD15 and BD20

Petaluma River E. San Pablo Bay (midpoint)

E. San Pablo Bay(nearshore)

Pacheco Creek Grizzly Bay

Site-Specific Objective Derivation March 200510

Figure 2a. Dissolved Copper WER Results by Sampling Station

1

2

3

4

5

6BA

40

BB15

LCB0

1

LCB0

2

BB30

BC10

BD20

SPB0

1

BD15

SPB0

2

SPB0

3

BF10

BF20

Diss

olve

d Cu

WER

Figure 2b. Dissolved Copper WER Results by Event

1

2

3

4

5

6

Event 1 Event 2 Event 3 Event 4 All but BD15

Dis

solv

ed C

u W

ER

N = 4 4 4 4 4 4 4 4 4 4 4 3 3

N = 13 13 13 11 46


3.2.2 Pooled WER Results

One goal of the NDB WER project was to determine whether significant spatial differencesexisted that would warrant having more than one WER and resultant SSO NDB. The graphicaland simple statistical review of the individual station and event results presented above did notshow strong evidence of spatial patterns. In terms of temporal variability, three of the four eventsshowed similar results and the fourth showed consistently elevated WERs, representingconditions when copper would be even less bioavailable.

To investigate the potential for spatial variability issue further, individual WER values from the13 sampling sites were pooled into six qualitative categories based on the available data. Thecategories and number of stations within each are shown below.

• Central Bay (6) • All sites except BD15• North Bay (7) • Shallow water sites (5)• All Sites (13) • Deep water sites (8)

The categories do not strictly mirror the hydrodynamic RMP redesignation segmentation or the“old” Basin Plan Bay segmentation. The Central Bay includes the sampling sites starting near theBay Bridge at Yerba Buena Island (BC10) and extending south of the San Mateo Bridge toRedwood Creek (BA40). A portion of the Lower Bay is thus included in the Central Baydesignation.

The North Bay grouping includes the sampling sites north and east of the San Pablo Bay Station(BD20) and roughly all areas upstream of the Richmond-San Rafael Bridge. The All Sitesgrouping includes all 13 Central and North Bay sites. The “All sites but BD15” grouping wasanalyzed to investigate to what extent the atypical results at the mouth of the Petaluma River(BD15) might skew the overall data set if included.

Shallow water (or mudflat) sites refer to the five new transect or “near-shore” sites that wereselected to investigate the existence of potential gradients from RMP “spine” stations towardsthe shore. Three such shallow water sites were included in the study in the North Bay and two inthe Central Bay. Deep water sites refer to the eight existing RMP spine stations included in theNDB WER study. The spine stations are in channelized areas of San Francisco Bay but are notnecessarily in physically deep water given the overall shallowness of the Bay. Four deep watersites were included in each of the North and Central Bays. The three physically deepest sites (30-40 feet) include Redwood Creek (BA40), Oyster Point (BB30) and Pacheco Creek (BF10). Theother sites were generally less than thirteen feet deep when sampled.

Summary statistics for pooled dissolved copper WERs are found in Table 3. Within each pooledgrouping there was a fairly even distribution of samples collected (e.g., 24 samples from theCentral Bay and 26 from the North Bay, or 20 shallow or near-shore water samples versus 30deep water of RMP spine samples).

When comparing statistics between these pooled groupings, it is evident that there is minimalvariability in all rows. For instance, the maximum WER values for all categories range from 5.2– 5.3. Similarly, the 5th percentile values range from 1.6 – 2.2. Central tendency WERs


(arithmetic mean, geometric mean and median) were quite consistent across all the groupingsevaluated with values between 2.4 and 2.8 These consistencies may indicate that a Bay-wideversus region specific WER would be appropriate. Additional statistical evaluation of spatial andtemporal variability is presented below.

Table 3. Dissolved Copper Pooled WER Summary Statistics

SummaryStatistics

CentralBay

NorthBay

AllSites

All butBD15

ShallowSites

DeepSites

number 24 26 50 46 20 30minimum 1.8 1.5 1.5 1.5 1.7 1.5maximum 5.2 5.3 5.3 5.2 5.2 5.3a. mean 2.8 2.5 2.7 2.6 2.6 2.7g. mean 2.7 2.4 2.6 2.5 2.5 2.690th Percentile 4.0 3.3 3.5 3.4 3.3 3.55th Percentile 2.2 1.6 1.7 1.7 1.7 1.6median 2.5 2.4 2.5 2.5 2.4 2.5Std. deviation 0.8 0.8 0.8 0.7 0.9 0.8

3.2.3 Additional WER Data Statistical Analyses

The NDB WER Report (July 2002 – Section 6.5) presented a more detailed statistical evaluationof the WER pooled results presented above to further evaluate the extent of variability andclustering of the data. Selected results from that evaluation are summarized below.

The results of the Kolmogorov-Smirnov goodness-of-fit tests and inspection of normalprobability plots conducted had indicated that the WER data were approximately normallydistributed. Therefore, no transformation of these data was necessary for subsequent repeatedmeasures of analysis of variance (ANOVA) tests and other statistical analyses.

Repeated measures ANOVA results showed that whether a site was shallow or deep had nosignificant effect on the WER. The effect of “Event” was not significantly different for shallowand deep sites. Mean WERs were found to vary significantly between events. Table 4a showsthat there was a significant difference between WERs at each site during all four sampling events(i.e., p-value < 0.05 indicates significant difference with 95% confidence). Event 4 had a lower95% confidence level value with a WER of 1.96, while Event 2 had a lower 95th % confidencevalue of 2.97.


Table 4. Mean WERs with 95% Confidence Intervalsa.

DF t-Value P-valueMeanWER

95% LowerConfidence

95% UpperConfidence

Event 1 12 12.258 <0.0001 2.25 2.03 2.47

Event 2 12 9.349 <0.0001 3.56 2.97 4.16

Event 3 12 12.302 <0.0001 2.58 2.30 2.86

Event 4 10 9.929 <0.0001 2.24 1.96 2.51

b.

Shallow 19 8.276 <0.0001 2.63 2.22 3.04

Deep 29 11.939 <0.0001 2.70 2.41 2.99

c.

Total 49 14.530 <0.0001 2.67 2.44 2.90

Plots illustrating mean WERs and 95% confidence intervals are provided below. Figure 3illustrates that there was no significant difference between deep and shallow mean WERs whendata were combined for all events. Figure 4 shows that there was little variation between deepand shallow WERs between events. However, there was some variability in mean WERsbetween events (i.e., higher WERs in Event 2). The pattern of variation was consistent for deepand shallow water sites. Figure 5 combined shallow and deep water site WERs into one meanfor each event and then for all events.

The mean WER for all events is 2.67 (Figure 4c). The lower 95 % confidence for this combinedWER is 2.44 and the upper 95 % confidence is 2.90.

Figure 3. Mean WERs: Deep versus Shallow

WER

1

1.5

2

2.5

3

3.5

deep shallow

LegendMean with 95% Confidence Limits


Figure 4. Mean WERs: Deep versus Shallow for Each Event

1

1.5

2

2.5

3

3.5

4

4.5

5

5.5

Event 1 Event 2 Event 3 Event 4

shallow

deep

Means with 95% Confidence Limits

WER

Figure 5. Mean WERs: Event (a.) and Overall (b.)

1

1.5

2

2.5

3

3.5

4

4.5

WERs1

1.5

2

2.5

3

3.5

4

4.5

Event 1 Event 2 Event 3 Event 4

Mean with 95% Confidence Limits

WER

ALL

3.2.4 Site-to-Site Variations

To extend the potential Bay WER segmentation analysis presented previously, the sites sampledin this study were also grouped into four areas instead of two, as indicated in Table 5 below.This grouping follows the historic Basin Plan segmentation (since superceded as part of the RMPredesign). A limitation of this pooling approach is that there are only four datapoints for theCentral Bay and 6 for Suisun Bay and 20 datapoints each for the Lower and San Pablo Baygroupings.

Table 5. Major Subsections of San Francisco Bay North of the Dumbarton Bridge

Lower Bay Central Bay San Pablo Bay Suisun BayBB30 LCB01 BC10 BD15 LCB01 BF10BB15 LCB02 BD20 LCB02 BF20BA40 LCB03

a. b.


When using this grouping approach, there is a slight but statistically significant differencebetween the “Lower Bay” and “San Pablo Bay” sites (p < 0.05). Table 6 shows p-values forcomparisons between each of the groupings. Figure 6 illustrates the different mean WERs ineach of the subsections. These results could be interpreted to indicate that it may be appropriateto compute separate WERs for the San Pablo Bay and Lower Bay areas of the NDB WER study.However, further comparison shows that the difference between these two areas is small and thatalternatively an average WER ± 0.5 could be considered for application to the entire Bay northof the Dumbarton Bridge. As shown in Table 4c, 0.5 is approximately the range between theupper and lower 95% confidence intervals for the Total (All Sites) pooled WER alternative.

Table 6. Subsection Comparisons and P-values

Site Comparisons: P-valueLower Bay – Central Bay 0.2691Lower Bay – San Pablo Bay 0.0004Lower Bay – Suisun Bay 0.7526Central Bay – San Pablo Bay 0.6301Central Bay – Suisun Bay 0.4234San Pablo Bay – Suisun Bay 0.3124

Figure 6. Mean WERs in Distinct Areas of San Francisco Bay

1

1.5

2

2.5

3

3.5

4

LowerBay

San PabloBay

Central Bay Suisun Bay

WER

Means with 95%

Confidence Limits


3.2.5 Sample Specific WER Results

An aspect of spatial variability not directly addressed by WER measurements involvesevaluating whether the measured ambient copper concentrations are exceeding toxicity thresholdvalues (Hypothesis H7 in the NDB WER report). However the WER data can be used in anindirect manner to evaluate this issue by conducting what the WER guidance describes a“sample-specific WER approach” [U.S. EPA, 1994].

In this approach, a quotient is calculated by dividing the concentration of dissolved copper (ateach station) for each event by the product of the national WQC (3.1 µg/L) times the WERobtained for each station.

The WER guidance states that “when the quotient for a sample is less than 1.0, the concentrationof the metal in that sample is acceptable, when the quotient for a sample is greater than 1.0, theconcentration of metal in that sample is too high [U.S. EPA, 1994].”

A table of these values using the NDB data showed that all such quotients were less than 1.0(Table 7). Similar results were submitted in 2001 and 2002 and part of the 2002 303(d) listingprocess to support the fact that the bay was not being impaired by ambient copperconcentrations.

Table 7. Sample-Specific WER Approach Results

Station Event 1 Event 2 Event 3 Event 4BA40 0.3 0.2 0.3 0.3BB15 0.4 0.2 0.2 0.3

LCB01 0.3 0.2 0.4 0.3LCB02 0.4 0.2 0.3 0.4BB30 0.3 0.2 0.2 0.2C

entr

al B

ay

BC10 0.3 0.2 0.2 0.2BD20 0.4 0.2 0.2 0.4SPB01 0.4 0.3 0.2 0.4BD15 0.5 0.3 0.3 0.5SPB02 0.5 0.2 0.3 0.5SPB03 0.5 0.3 0.2 0.5BF10 0.4 0.2 0.2 *

Nor

th B

ay

BF20 0.5 0.3 0.4 **data did not meet QA/QC criteria and were not used in WER analyses

Measured Cu (µg/L)

(3.1 µg/L)(Cu WER)


3.3 San Jose Approach to Final WER and SSO Selection

This section reviews the approach and reasoning used by the City of San Jose in their WERstudy (May 1998) in evaluating final WERs (FWERs) and calculating alternative SSOs for thebay SDB. The discussion below focuses on the reasoning employed when evaluating the prosand cons of different data (station) pooling alternatives. Many of the variables assessed by SanJose for SDB are relevant to the FWER decisions to be made NDB.

3.3.1 FWER Selection

The San Jose study used both total and dissolved copper WER results from three stations(Dumbarton North, Dumbarton South and Coyote Creek) in their development and ultimateselection of a final Water Effects Ratio and calculation of a SSO. {Based on the greatervariability in total WER results and the U.S. EPA support (since 1993) for dissolved WQOs,total WERs were calculated for and included in the WER report Appendices but were notconsidered for adoption as part of the NDB WER study}.

For San Jose, results from the two Dumbarton stations were very similar for all aspects of waterquality, toxicology and determined WERs. The significant differences in WER values betweenthe northern (two Dumbarton stations) and southern (Coyote Creek station) portions of the SouthBay suggested that two site-specific criteria could be applied to the South Bay. For instance, ahigher criterion based only on Coyote Creek WERs could be established for the southernmostportion of the South Bay (nearer to POTW flows) and a lower criterion established for thenorthern portion of the South Bay based on WERs from the Dumbarton Bridge stations. Aconcern about this multi-WER alternative was that it may have needed additional supportinginformation (e.g., dilution modeling involving all three POTWs in the South Bay). Nevertheless,it recognized the unique water quality characteristics and protection provided by differentportions of the Bay, a factor to be considered in setting appropriate (i.e. neither under protectivenor overprotective) site-specific criteria in the Bay.

In recognition of the regulatory complexities associated with a multiple SSO approach, twoalternatives were developed by San Jose for the derivation and use of a single FWER value forthe South Bay. These were a three-station pooled FWER (n=60) and a two-station pooled(Dumbarton) FWER (n=40). The uncertainty, albeit small, associated with the three-stationFWER’s protectiveness at the northern end of the study site led San Jose to suggest use of thetwo-station FWER of 2.771 to determine a site-specific criterion versus the three-station WER of3.005.

3.3.2 Recalculation of the National Copper Criterion

A site-specific criterion is the product of the selected FWER and the national criterion (NationalCriterion * FWER = Site-Specific Criterion). The WER guidance (U.S. EPA, 1994a) suggeststhat the national criterion should first be evaluated and, as appropriate, modified using suitablequality site-specific data, prior to calculating the site-specific criterion. The current nationalsaltwater copper Criterion Maximum Concentration (CMC) and Criterion ContinuousConcentration (CCC) are 4.8 and 3.1 µg/L dissolved copper, respectively (U.S. EPA, 1995a).Prior to using the national criterion in the calculation, San Jose first recalculated it based uponthe new information provided by the results from its study. The new data consisted of three


EC50 values for Strongylocentrotus purpuratus and six new EC50 values for M. edulis. Usingthe new data and following the appropriate national criteria derivation process (U.S.EPA, 1985),modified national criteria (CMC & CCC) were produced.

The current national saltwater copper Final Acute Value (FAV) is 10.39 µg/L based on the fourmost sensitive species. However, this FAV was lowered to 9.625 µg/L, the Species Mean AcuteValue (SMAV) for M. edulis, in order to protect this commercially important species pursuant toUSEPA guidance. As a result, the current national saltwater copper CMC is 4.8 µg/L(9.625/2=4.8). The current national saltwater copper CCC is 3.1 µg/L, which is the quotient ofthe SMAV of 9.625 µg/L and the current (U.S. EPA, 1995c) Acute/Chronic Ratio of 3.127(9.625/3.127=3.1). This SMAV is derived from four SAIC (1993) and three ToxScan (1991a, b& c) values (Table 8).

When the six reference toxicant test dissolved copper EC50 values from the San Jose study wereadded to the above seven values, the new SMAV for M. edulis decreased from 9.625 µg/L to7.888 µg/L (Table 8). This resulted in a new CMC of 3.9 µg/L dissolved copper (7.888/2=3.9)and a new CCC of 2.5 µg/L dissolved copper (7.888/3.127=2.5; Table 8).

This San Jose recalculation of the national copper criteria (CMC & CCC) was intended toproduce conservative, scientifically defensible WER results. Whether this approach andpermanent inclusion of the San Jose data (for M. edulis and S. purpuratus) into the nationalcopper database is appropriate for other copper WER and SSO studies is subject to additionalregulatory review.

3.3.3 San Jose SSO Selection

The U.S. EPA procedure and formula for calculating site-specific criteria (National Criterion *WER = Site-Specific Criterion) were used to calculate site-specific CCC values for the LowerSouth Bay WER study. The San Jose results supported either a two-station or a three-stationFWER in deriving an appropriate site-specific CCC for the South Bay. The three-stationdissolved FWER, based on the geometric mean of corrected dissolved WER values from allthree stations in the study site (n=60) was 3.005. The two-station dissolved FWER, based on thegeometric mean of corrected dissolved WER values for the two Dumbarton stations (n=40) was2.771. Multiplying the three-station FWER by the modified national CCC of 2.5 µg/L produceda site-specific CCC of 7.5 µg/L dissolved copper. Multiplying the two-station FWER by themodified national CCC produced a site-specific CCC of 6.9 µg/L dissolved copper.

There were significant differences in WER values between the two Dumbarton Bridge stationsand the Coyote Creek station. Therefore, this criterion (7.5 µg/L) would simultaneously under-protect the northern portion of the site (Dumbarton Bridge CCC = 6.9 µg/L) and overprotect thesouthern portion of the site (Coyote Creek CCC = 8.8 µg/L). This simultaneous overprotectionand under-protection reflected an inherent drawback of implementing a single site-specificcriterion in a site where the data demonstrated the potential need for a multiple criteria approach.

The pooled two-station Dumbarton Bridge site-specific CCC of 6.9 µg/L was the valueultimately supported by the TMDL workgroup and adopted by the RWQCB into the Basin Plan.


Table 8. Current and Modified National Copper Criteria to reflect the addition of NDBWER study data into the San Jose study modified EPA National Copper Criteria database.

Dissolved Copper Criteria*

Data Source Proposed NationalCriteria (EC50) Data

(USEPA 1995a)

Six San Jose (1998) EC50Values Plus Proposed NationalCriteria Data (USEPA 1995a)

National Criterion Data, SanJose Data Plus Eight NDB

Values (July 2002)SAIC (1993) 12.5 12.5 12.5SAIC (1993) 14.1 14.1 14.1SAIC (1993) 11.3 11.3 11.3SAIC (1993) 11.9 11.9 11.9ToxScan (1991a) 5.787 5.787 5.787ToxScan (1991b) 8.889 8.889 8.889ToxScan (1991c) 6.278 6.278 6.278San Jose (1998) 5.024 5.024San Jose (1998) 4.392 4.392San Jose (1998) 7.497 7.497San Jose (1998) 6.789 6.789San Jose (1998) 6.822 6.822San Jose (1998) 7.806 7.806NDB Study 8.05NDB Study 8.32NDB Study 5.64NDB Study 9.36NDB Study 7.08NDB Study 6.91NDB Study 6.80NDB Study 9.44FAV 9.625 7.888 7.776CMC 4.8 3.9 3.9ACR 3.127 3.127 3.127CCC 3.1 2.5 2.5FAV = Final Acute ValueCMC = Criterion Maximum ConcentrationACR = Currently proposed Acute-Chronic Ratio (U.S. EPA, 1995c)CCC = Criterion Continuous ConcentrationCMC = FAV/2; CCC = FAV/ACR* The proposed national copper criteria are based solely on results with Mytilus edulis in order toprotect this commercially important species.

3.4 NDB Final WER and SSO Selection

The current copper WQO applicable to San Francisco Bay NDB is the CTR CCC value of 3.1µg/L times a WER (the default WER is 1.0). Table A1 in Appendix A shows the complete set ofindividual NDB calculated site-specific WER based SSOs for each of the four events at eachsampling station. These are the copper objective alternatives that are directly sanctioned by theCTR.

For comparative purposes, also shown are the equivalent individual station SSOs derived fromthe WERs multiplied by the San Jose (and NDB) adjusted national criterion of 2.5 µg/L. As


shown in Table 8, adding in the eight labwater samples from the NDB study to the combinedSan Jose (1998) and national criterion dataset, produced the same recalculated CCC value of 2.5µg/L as derived using the San Jose and national dataset.

The CTR WQO based SSOs for all four events ranged from 5.2 µg/L (BF20) to 8.4 µg/L (BA40and BD15). While some of the ambient copper values approached the non-WER adjusted 3.1µg/L level, most would be a factor of two to three below a SSO based on the WERs developed inthe NDB study and either the CTR or recalculated WQOs (Table A1).

3.4.1 Comparison of North Of Dumbarton Study and San Jose Study WER Results

Results of the north of Dumbarton study are quite consistent with results obtained during the1996-1997 San Jose study (Table 9). The Redwood Creek station (BA40) was investigated inboth studies (in 2000 – 2001 and 1996 – 1997) and results were comparable (averages of 2.75and 2.2). The City of San Jose used the BA40 results for comparative purposes but not forcalculation of a final WER. Lab water results from the two studies were also in agreement,supporting the validity of comparing the two studies [see WER report and Table 8].

Table 9. Comparison of North of Dumbarton and San Jose Dissolved Copper WER Results

North of Dumbarton StudySummaryStatistics

CentralBay

NorthBay

AllStations

All StationsExcept BD15

ShallowStations

DeepStations

Number 24 26 50 46 20 30Minimum 1.8 1.5 1.5 1.5 1.7 1.5Maximum 5.2 5.3 5.3 5.2 5.2 5.3a. mean 2.8 2.5 2.7 2.6 2.6 2.7g. mean 2.7 2.4 2.6 2.5 2.5 2.6Std. Deviation 0.8 0.8 0.8 0.7 0.9 0.8

San Jose Study

Summary StatisticsSan

Mateo

North ofDumbarton

Bridge

South ofDumbarton

Bridge

CoyoteCreek

Number 7 20 20 20Minimum 1.7 2.2 2.2 2.5Maximum 2.4 3.9 4.5 4.8a. mean 2.2 2.7 2.9 3.6g. mean 2.1 2.7 2.8 3.5Std. Deviation 0.3 0.4 0.6 0.8


3.4.2 NDB Site-Specific CCC (SSO) Recommendation

A primary goal of the NDB WER study was to produce scientifically defensible WER valuesthat could be used with confidence by State and U.S. EPA regulators, dischargers andstakeholders to establish one or more SSOs for the Bay north of Dumbarton Bridge. Severalconservative measures were employed in both studies including: using M. edulis, the mostsensitive species listed in the marine criteria data set for copper, as the test species; andconsideration of lowering the national CCC from 3.1 to 2.5 µg/L dissolved copper byincorporating the site-specific laboratory water results into the national copper data set.

As shown in Table 9 there was a relatively small variation in the WERs for the different poolingalternatives. The statistical analysis had shown the WER data to be normally distributed. TheUSEPA guidance suggests using geometric means for FWER selection. The arithmetic meansranged from 2.5 to 2.8 while the geometric means ranged from 2.4 to 2.7. The All Sitesarithmetic and geometric means are 2.7 and 2.6, respectively, in the middle of the alreadyrelatively narrow Central to North Bay range cited. The prior statistical analysis had shown thereto be no significant differences between results at shallow versus deep water stations so thosegroupings are not considered further in the SSO selection analysis.

Site-specific CCC (SSO) values and associated summary statistics were calculated and shownbelow (Table 10) for the each of the four different sets of pooled station WER data. Forcomparative purposes, results are shown for SSOs derived from both the currently applicableCTR WQO (CCC) value of 3.1 µg/L and the San Jose and NDB data recalculated national valueof 2.5 µg/L (Table 8). The prior statistical analysis found only a minor difference in WERs (0.5)between a pooling of Lower Bay versus San Pablo Bay stations. It further found that 0.5 wasapproximately the difference between the upper and lower 95% confidence intervals for the AllSites pooled WER alternative. Additional analysis showed there to be no statistically significantdifference between the Central Bay and the North Bay pooled WERs.

These relatively small differences between the various pooled dataset provides some support forselecting a single NDB SSO using all the available data. If arithmetic averages are used (giventhat the data are normally distributed), an All Sites NDB SSO could range from 6.8 to 8.4 µg/L,depending on whether the CTR or recalculated WQO is used as the basis of adjustment by theAll Sites FWER of 2.7.


Table 10. NDB Pooled Station Dissolved Copper SSO Summary Statistics

North Bay Central Bay All Sites All but BD15Summary Statistics CTR

SSORecalcSSO

CTRSSO

RecalcSSO

CTRSSO

RecalcSSO

CTRSSO

RecalcSSO

Minimum 5.6 4.5 4.7 3.8 4.7 3.8 4.7 3.85th Percentile 6.8 5.5 5.0 4.0 5.3 4.3 5.3 4.3Median 7.8 6.3 7.4 6.0 7.8 6.3 7.8 6.4a. mean 8.7 7.0 7.8 6.3 8.4 6.8 8.1 6.5g. mean 8.4 6.8 7.4 6.0 8.1 6.5 7.8 6.390th Percentile 12.4 10.0 10.2 8.3 10.9 8.8 10.5 8.5Maximum 16.1 13.0 16.4 13.3 16.4 13.3 16.1 13.0

ArithmeticMeans

NorthBay

CentralBay

AllSites

All Sitesbut BD15

CTR SSO 8.7 7.8 8.4 8.1Recalc SSO 7.0 6.3 6.8 6.5

The low end of All Sites range is close to the Lower South Bay SSO value of 6.9 µg/L. Thecopper speciation results (Bruland 2003) provide an independent line of evidence that a SSOvalue in that range could be considered protective bay-wide. The report calculated that if theconcentration of dissolved copper increased to a value of 6.9 µg/L, or 108 nM, it would raise theambient [Cu2+] to 10-11 M, a concentration that could impair the health and viability of theplankton. The study found that strong copper-complexing ligands dominate the chemicalspeciation of dissolved copper throughout San Francisco Bay, including the Central Bay (YerbaBuena Island station). The concentrations of these ambient organic ligands exceeded the totaldissolved copper concentrations at every site, and these ligands complexed greater than 99.9% ofthe dissolved copper. Regardless of site or season, the [Cu2+] values throughout San FranciscoBay did not exceed 10-13 M, a value deemed to be suitably below the toxicity limit for aquaticorganisms.

3.4.3 Revised SSO Recommendation

Since the time of the WER study, the RMP has reevaluated the regional definitions in the Bay.The RMP now recognizes 5 regions (see Figure 7):

1. Suisun Bay2. San Pablo Bay3. Central Bay4. South Bay5. Lower South Bay


Figure 7. RMP Newly Defined Regions of San Francisco Bay

Rather than keeping the SDB work separate from the NDB work, it has been found that it ismore appropriate to integrate the studies and create two SSOs for the entire Bay. These potentialdissolved copper SSOs are calculated as 6.0 ppb for Bay Regions 1-3 and 6.9 ppb for BayRegions 4 & 5. This approach protects Mytilus sp., the most sensitive species in the EPAdatabase and a commercially important species. These SSOs are the result of two proposed WERvalues (2.4 for Regions 1-3; 2.7 for Regions 4-5). San Bruno Shoal was identified as the linebetween Regions 3 and 4. Further discussions on this approach can be found in Appendices Cand D.

4. BIOTIC LIGAND MODELThe Biotic Ligand Model (BLM) predicts metal toxicity to aquatic organisms based on thechemical characterization of a given water body. The model takes into consideration severalwater quality parameters, including hardness, DOC, chloride, pH, and alkalinity. The BLM wasused to predict copper toxicity in the NDB WER study water samples from San Francisco Bay,and in the laboratory water samples from the Granite Canyon Marine Laboratory in Carmel, CA.

LowerSouth Bay

Bay Bridge

N

SouthBay

12

3

4

5

CentralBay

San PabloBay

SuisunBay

GoldenGate


BLM input chemistry was measured in the second, third, and fourth sampling events. This modelwas previously developed with toxicity data from South San Francisco Bay WER study (Paquinet al., 2000). The model was used to predict EC50s “blind”, i.e. without knowing the measuredvalues until after predictions were made.

Previous comparisons of BLM predictions with measured toxicity data have established plus orminus a factor of 2 as a standard comparison indicating good agreement between predicated andmeasured values (Santore et al., 2001). Comparison of measured and predicted EC50 values inthese results initially showed a number of samples that fell outside the plus or minus a factor of 2zones that indicates good agreement (Figure 8 upper). The DOC concentration was reported asless than detection for many of these samples where the BLM predicted EC50s significantlydifferent than measured values. Comparison of the measured DOC concentrations for each roundof samples taken from a given station shows that these below detection limit values areanomalously low for these samples. Additionally, a reported DOC concentration of 11 mg/L fora Granite Canyon seawater sample appears to be anomalously high with a DOC concentrationmuch higher than any of the field samples.

All of these samples with very low or high DOC concentrations were also cases where the BLMpredictions based on those reported DOC concentrations did not match well with measuredvalues. When these suspect TOC samples are censored, the comparison of predicted andmeasured Cu EC50s improves dramatically (Figure 8 middle). As discussed in the July 2002WER report, two North Bay Event 4 toxicity test results were determined to be unreliable andnot used is the WER calculations. When these values are censored (Figure 8 bottom) the modelpredictions improve further, with only 1 out of 58 samples falling slightly outside the range ofgood agreement.

There does appear to be a systematic bias in the model results showing predictions that are toolow at DOC concentrations below 4.0 mg C/L, and too high at DOC concentrations above thisvalue. In general, the Cu BLM applied to estuarine data appears to predict too strong an impactof DOC on copper toxicity. This discrepancy has not been seen in freshwater datasets. TheUSEPA is currently reviewing the BLM as a potentially less resource intensive option to WERstudies for the development of site-specific criteria.

Overall, the BLM results provided an independent confirmation of the high quality and reliabilityof the toxicity test data used to develop the NDB copper WERs and resultant SSOs.


Figure 8. Comparison of BLM Predicted versus NDB Study Measured Toxicity (MytilusEC50)

All EC50 Data

1

10

100

1 10 100Measured EC50 (ug/L)

Pred

icte

d EC

50

(u

g/L)

GC LabwaterCentral Bay

North Bay

BLM Over Predicts Toxicity

BLM Under Predicts Toxicity

No Event 4 Lab Water or Central Bay Sites

1

10

100


Pred

icte

d EC

50

(u

g/L) GC Labwater

Central BayNorth Bay



No Event 4 Lab Water, Central Bay Sites or BF10 & 20

1

10

100


Pred

icte

d EC

50

(u

g/L) GC Labwater

Central Bay

North Bay




5. COMPLIANCE EVALUATION WITH SSO BASEDEFFLUENT LIMITSThe SIP SSO Justification Request Report (Draft February 27, 2004) included an evaluation ofthe ability of three case study POTWs to comply with non-WER adjusted CTR based copper andnickel effluent limits. Tables 11 to 14 present similar compliance evaluations except witheffluent limits calculated using a range of copper SSOs developed in this report based on therange of arithmetic and geometric mean WERs for the pooled North and Central Bay stations(2.4 – 2.8). These effluent limits use the associated pooled North and Central Bay median and90th percentile translators (depending on discharge location) presented in the companion to thisreport the Translator Development and Derivation Report (Draft March 12, 2004).

Copper compliance continues to be an issue for shallow water (zero dilution) municipalsecondary treatment plants such as LGVSD no matter what WER/SSO is selected (Table 11).Copper compliance may also continue to be an issue also for shallow water advanced secondaryplants such as FSSD, depending on the SSO selected. Deepwater secondary treatmentdischargers (with 10:1 dilution) with performance equivalent to EBMUD would appear to haveminimal compliance issues with any SSO based limit.

Compliance with any of the SSO based nickel effluent limits does not appear to be an issue withthese POTWs (Table 12).

Tables 13 and 14 show the projected percentage of all dischargers to comply with the range ofSSO based effluent limits as read from the probability plots generated from the pooled ERS 2001– 2003 effluent dataset (see the SIP SSO Justification Report). Secondary treatment POTWs andindustries without dilution credit will have moderate to significant copper compliance problemseven with the upper range of SSO based effluent limits. Advanced secondary POTWs withoutdilution may have minor compliance problems if relatively low WER based SSOs are selected. Asmall percentage of facilities with 10:1 dilution may have copper compliance problems ifrelatively low copper WER based SSOs are selected (Table 13).

A small percentage of industries without dilution credit may have compliance problems witheffluent limits derived from the low end of the SSO range (Table 14).


Table 11. Copper SSO Based Effluent Limits – Case Study Compliance Evaluation

LGVSD FSSD EBMUD

WERChronicSSO(µg/L)

AMEL(µg/L)

MECComply?(25 µg/L)

99.87%Comply?

(31.8 µg/L)

AMEL(µg/L)

MECComply?(9.0 µg/L)

99.87%Comply?

(10.8 µg/L)

AMEL(µg/L)

MECComply?

(25.9 µg/L)

99.87%Comply?

(35.8 µg/L)

1.0 2.5 3.5 No No 3.8 No No 2.9 No No

1.0 3.1 4.4 No No 4.6 No No 7.6 No No

2.4 6.0 8.5 No No 9.0 Yes No 40.2 Yes Yes

2.4 7.4 10.4 No No 11.1 Yes Yes 53.6 Yes Yes

2.8 7.0 9.9 No No 10.5 Yes No 49.9 Yes Yes

2.8 8.7 12.2 No No 12.9 Yes Yes 65.5 Yes YesNote: AMELs assume use of NDB pooled North Bay or Central Bay metals translators and total metals ambient concentrations. EBMUD 10:1 dilution.

Table 12. Nickel SSO Based Effluent Limits – Case Study Compliance Evaluation

LGVSD FSSD EBMUD

WER

ChronicSSO

(µg/L)

AMEL(µg/L)


99.87%Comply?

(10.8 µg/L)

AMEL(µg/L)


99.87%Comply?(8.6 µg/L)

AMEL(µg/L)

MECComply?

(16.0 µg/L)

99.87%Comply?

(16.7 µg/L)

1.0 8.2 27.8 Yes Yes 27.8 Yes Yes 82 Yes Yes



1.0 20.9 70.8 Yes Yes 70.8 Yes Yes 227 Yes YesNote: AMELs assume use of NDB pooled North Bay or Central Bay metals translators and total metals ambient concentrations. EBMUD 10:1 dilution.


Table 13. Copper SSO Based Effluent Limits – All Dischargers Compliance Evaluation

Shallow Water Dischargers Compliance(zero dilution)

Deepwater Dischargers Compliance(10:1 dilution)

POTW % POTW %WER

ChronicSSO

(µg/L)

AMEL(µg/L) Secondary Adv. Secondary

Industry%


Industry%

1.0 2.5 3.5 6 45 20 2.9 4 33 15

1.0 3.1 4.4 12 57 30 7.6 40 89 55

2.4 6.0 8.5 50 94 58 40.2 >99.9 99.8 98

2.4 7.4 10.4 63 97 65 53.6 >99.9 >99.9 98.4

2.8 7.0 9.9 60 97 60 49.9 >99.9 >99.9 98.3

2.8 8.7 12.2 75 98 72 65.5 >99.9 >99.9 98.8Note: AMELs assume use of NDB pooled North Bay or Central Bay metals translators and total metals ambient concentrations. EBMUD 10:1 dilution.

Table 14. Nickel SSO Based Effluent Limits – All Dischargers Compliance Evaluation

Shallow Water Dischargers Compliance(zero dilution)

Deepwater Dischargers Compliance(10:1 dilution)

POTW % POTW %WER

ChronicSSO

(µg/L)


Industry%


Industry%

1.0 8.2 27.8 99.7 >99.9 92 82 >99.9 >99.9 99.5

1.0 11.9 40.3 >99.9 >99.9 98 132 >99.9 >99.9 >99.9

1.0 16.4 55.5 >99.9 >99.9 99.2 194 >99.9 >99.9 >99.9

1.0 20.9 70.8 >99.9 >99.9 99.4 227 >99.9 >99.9 >99.9Note: AMELs assume use of NDB pooled North Bay or Central Bay metals translators and total metals ambient concentrations. EBMUD 10:1 dilution.


6. REFERENCES

Carlson et al. 1984. Guidelines for deriving Numerical Aquatic Site-Specific Water QualityCriteria by Modifying National Criteria. USEPA (EPA-600/3-84-099).

EOA, Inc. and LWA, Inc, 2000. Work Plan for Copper and Nickel Impairment Assessment toAssist in Preparation of 2002 303(d) List: San Francisco Bay, North of Dumbarton Bridge.Prepared for North Bay Dischargers Group, Bay Area Dischargers Association and WesternStates Petroleum Association, August 17, 2000.

EOA, Inc. and LWA, Inc., 2002. Copper & Nickel North of the Dumbarton Bridge. Step 1:Impairment Assessment Report, Ambient Concentrations and WERs, September 2000 – June2001.

EOA, Inc. and LWA, Inc., 2004. Clean Estuary Partnership: North of Dumbarton Bridge Copperand Nickel Conceptual Model and Impairment Assessment Report. December.

EOA, Inc. and LWA, Inc., 2004. Clean Estuary Partnership: North of Dumbarton Bridge Copperand Nickel Development and Selection of Final Translators. December.

EOA, Inc. and LWA, Inc., 2004. Clean Estuary Partnership: North of Dumbarton Bridge Copperand Nickel Site-Specific Objectives State Implementation Policy Justification Report. December.

SAIC. 1993. Toxicity testing to support the New York/New Jersey Harbor site-specific coppercriteria study - final report. EPA Contract No. 68-C8-0066. Science Applications InternationalCorporation, Narragansett, RI.

San Jose, City of. Environmental Service Department. 1998. Development of a Site-SpecificWater Quality Criterion for Copper in South San Francisco Bay.

ToxScan. 1991a. Results of provision E5F spiked metals toxicity testing -- 14 to 21 February1991. Prepared for Kinnetic Laboratories, Inc. for "Site-Specific Water Quality Objectives forSouth San Francisco Bay" report by Larry Walker Associates and Kinnetic Laboratories undersubcontract to CH2M HILL. April, 1991.

ToxScan. 1991b. Results of provision E5F spiked metals toxicity testing -- 27 February to 6March 1991. Prepared for Kinnetic Laboratories, Inc. for "Site-Specific Water QualityObjectives for South San Francisco Bay" report by Larry Walker Associates and KinneticLaboratories under subcontract to CH2M HILL. Revised July, 1991.

ToxScan. 1991c. Results of provision E5F spiked metals toxicity testing -- 2 to 9 April 1991.Prepared for Kinnetic Laboratories, Inc. for "Site-Specific Water Quality Objectives for SouthSan Francisco Bay" report by Larry Walker Associates and Kinnetic Laboratories undersubcontract to CH2M HILL. Revised July, 1991.


USEPA. 1985. Guidelines for Deriving Numerical National Water Quality Criteria for theProtection of Aquatic Organisms and their Uses. PB85-227049.

USEPA. 1994. Interim Guidance on Determination and Use of Water-Effects Ratios for Metals.EPA 823-B-94-001.

USEPA. 1995a. Method 1669: Sampling Ambient Water for Trace Metals at EPA WaterQuality Criteria Levels. EPA 821-R-95-034.

USEPA. 1995b. Short-Term Methods for Estimating the Chronic Toxicity of Effluents andReceiving Waters to West Coast Marine and Estuarine Organisms. EPA 600-R-95-136.

USEPA. 1995c. Ambient Water Quality Criteria - Saltwater Copper Addendum.

Watson et al. 1996. Recalculation of the National Marine Water Quality Criterion for Nickel andDevelopment of a Site-Specific Criterion. City of San Jose ESD, San Jose, CA and Tetra Tech,Inc. Owings Mills, MD.

Watson et al. 1999. Nickel Acute-to-Chronic Ration Study. City of San Jose water PollutionControl Plant. Environmental Service Department.

Appendix A

Individual Sample Station Calculated SSOsCompared to Ambient Dissolved Copper Concentrations

Site-Specific Objective Derivation March 2005A-1

Table A1. Comparison of Individual Station Calculated Site-Specific Dissolved Copper Water Quality Objectives withAmbient Dissolved Copper Concentrations. SSO Results Shown Based on both Existing CTR 3.1 µg/L WQO and AdjustedWQO of 2.5 µg/L

Event 1 - September (dry) Event 2 - February (wet) Event 3 - April (spring) Event 4 - June (dry)CTRSSO

RecalcSSO

Diss.Cu

CTRSSO

RecalcSSO

Diss.Cu

CTRSSO

RecalcSSO

Diss.Cu

CTRSSO

RecalcSSO

Diss.CuStation WER

µg/LWER

µg/LWER

µg/LWER

µg/LBA40 2.7 8.4 6.8 2.9 4.2 13.0 10.5 2.7 2.7 8.4 6.8 2.5 3.1 9.7 7.8 2.9BB15 2.4 7.5 6.0 2.9 3.2 10.0 8.0 2.1 2.7 8.3 6.8 2.1 2.5 7.8 6.3 2.0

LCB01 2.5 7.8 6.3 2.5 4.7 14.4 11.8 2.7 2.4 7.6 6.0 2.8 2.4 7.4 6.0 2.5LCB02 2.4 7.5 6.0 2.8 5.2 16.1 13.0 3.0 2.8 8.6 7.0 2.8 2.2 6.7 5.5 2.5BB30 2.5 7.8 6.3 2.6 3.5 10.8 8.8 2.2 2.4 7.4 6.0 1.6 2.4 7.5 6.0 1.7C

entr

al B

ay

BC10 2.2 6.9 5.5 1.9 2.6 8.0 6.5 1.3 2.4 7.4 6.0 1.3 1.8 5.7 4.5 1.3

BD20 2.2 6.8 5.5 2.5 2.6 7.9 6.5 1.9 2.0 6.2 5.0 1.5 1.5 4.8 3.8 2.1SPB01 2.0 6.2 5.0 2.5 2.6 8.1 6.5 2.4 2.9 9.0 7.3 1.8 2.0 6.3 5.0 2.5BD15 2.7 8.4 6.8 4.2 5.3 16.5 13.3 4.3 3.4 10.5 8.5 3.6 2.4 7.5 6.0 3.8SPB02 1.7 5.3 4.3 2.8 3.2 9.9 8.0 2.0 2.4 7.4 6.0 1.9 2.2 7 5.5 3.4SPB03 1.7 5.4 4.3 2.8 2.5 7.6 6.3 2.0 2.7 8.3 6.8 2.0 2.1 6.5 5.3 3.4BF10 2.5 7.9 6.3 2.8 3.5 10.9 8.8 2.5 3.1 9.6 7.8 2.3 * * * 2.7

Nor

th B

ay

BF20 1.7 5.2 4.3 2.8 3.2 9.9 8.0 2.6 1.6 5.0 4.0 2.2 * * * 2.3

CTR SSO = WER X 3.1 µg/LRecalc = WER X 2.5 µg/L*Data did not meet QA/QC criteria and were not used in calculations

Appendix B

Executive Summary toBay Area Clean Water Agencies

Copper & Nickel North of the Dumbarton BridgeStep 1: Impairment Assessment Report

Ambient Concentrations and WERsSeptember 2000 – June 2001

Site-Specific Objective Derivation March 2005B-1

EXECUTIVE SUMMARY

Introduction

In accordance with Section 303(d) of the Clean Water Act, States are required to list waters thatwill not comply with adopted water quality objectives after imposition of technology-basedcontrols on point source discharges. San Francisco Bay was listed on the 1998 303(d) list forCalifornia due to levels of total recoverable copper and nickel which exceeded 1986 Basin Plantotal recoverable metals objectives and/or USEPA national criteria. These exceedances were thebasis for a concern that copper and nickel were impairing aquatic uses in the Bay by producingeither acute or chronic toxicity in sensitive aquatic organisms.

Events have occurred since the 1998 listing, which have given rise to re-evaluation of the listing.In the California Toxics Rule (CTR) of 2001, new water quality objectives for copper and nickelwere adopted. Those objectives are based on the dissolved forms of copper and nickel,consistent with USEPA national policy, and provide for site-specific adjustments. Also, in SouthSan Francisco Bay, work performed by the City of San Jose has indicated that modification ofthe CTR objectives for copper and nickel is appropriate. The studies in South Bay haveindicated that 303(d) listing for copper and nickel is not appropriate.

To assess whether the 303(d) listings for copper and nickel in the rest of San Francisco Bay(north of the Dumbarton Bridge) should be modified or eliminated, it was recognized thatcomplementary scientific work to that conducted by the City of San Jose should be conducted.

A bay-wide stakeholder group (Coordinating Committee, CC) consisting of regulators, municipaland industrial dischargers, and environmental group members was assembled to overseedevelopment and implementation of a Work Plan that is consistent with the South Bay technicalapproach [Work Plan, 2000].

The primary purpose of the study outlined in the Work Plan was to collect data to improve theunderstanding of the aquatic toxicity of copper and nickel in San Francisco Bay north of theDumbarton Bridge. The study was designed to provide information useful to the State inpreparing the year 2002 303(d) list for San Francisco Bay. To meet this objective, the study wasdesigned (1) to provide data which is scientifically defensible (accurate, reproducible, etc.), (2)to provide data which fairly characterizes existing ambient water column levels of copper andnickel in San Francisco Bay north of the Dumbarton Bridge, (3) to provide data which will beuseful in the development of a site-specific water quality objective for copper for San FranciscoBay north of the Dumbarton Bridge, and (4) to provide data that will be useful in the derivationof “translator” values (relating dissolved and total ambient water column concentrations) whichare used in deriving NPDES permit limits for copper and nickel.

The Work Plan for this project included convening a Technical Review Committee (TRC) toprovide an independent outside critique of the project design and results.

This project has included participation from members of the following groups since its inception:North Bay Dischargers Group (NBDG), Bay Area Clean Water Agencies (BACWA), the


Western States Petroleum Association (WSPA), Bay Area Association of StormwaterManagement Agencies (BAASMA), San Francisco BayKeeper (SF BayKeeper), Regional WaterQuality Control Board staff (RWQCB), US Environmental Protection Agency staff (USEPA),San Francisco Estuary Institute (SFEI) and the Copper Development Association (CDA).

Sampling Procedures

Sampling was conducted at thirteen stations selected by the Coordinating Committee (CC) anddescribed in the August 17, 2000 Work Plan (Work Plan). In December 2000, the technicalreview committee (TRC) for this study met to review the Work Plan and results of the firstsampling run. As a result of the TRC meeting and subsequent Coordinating Committeediscussion, it was decided to complete the study using the original thirteen north of DumbartonBridge stations sampled during Event 1. Sample site selection was based on existing RMP data,results from hydrodynamic modeling, and the need to explore shallow areas of the Bay. Sampleevents included 8 RMP sample sites (located in main channels of the Bay) and 5 shallow watersites (located in mudflat areas) sampled over a two-day period. The shallow water sites werechosen to create transects anchored on deep water RMP sites, in order to develop information onpossible gradients extending into the shallows.

Sampling events were conducted during the period from September 2000 to June 2001. The goalof the sampling and toxicity testing was to produce four WER events (two from the dry seasonand two from the wet season). The rationale behind the sampling event selection was to capturethe dominant hydrological conditions observed during the year. The selected number of eventsalso represented a balancing of temporal coverage with the need for extensive spatial coverage toaddress representative areas of the Bay north of Dumbarton, both deep water and shallow water.

Clean sampling techniques were used for all fieldwork. All tubing and sample containers usedfor the collection of ambient water samples were cleaned following USEPA guidelines.

Laboratory Procedures

Laboratory tests used in the study included bioassay testing and chemical testing.

Mytilus edulis is the ideal organism for use in copper bioassays needed to determine WaterEffect Ratio (WER) values due to its sensitivity to copper. WER values are used to establishsite-specific adjustments to copper objectives, per the CTR. The Mytilus edulis toxicity testused for this study followed the guidelines established by the USEPA manual Short-TermMethods for Estimating the Chronic Toxicity of Effluents and Receiving Waters to West CoastMarine and Estuarine Organisms (EPA/600/R-95/136).

Chemical analyses performed for this study followed USEPA clean techniques and procedures.

Quality Assurance/Quality Control (QA/QC)

The objective of this QA/QC analysis was to assess the acceptability of data generated during thefour sampling events. Holding times, analytical accuracy and precision, potential contamination,and conformance to data acceptability criteria were investigated to determine if results needed


qualification. Furthermore, any questionable results or missing data were identified andinvestigated.

Analytical chemistry accuracy and precision were monitored throughout the four samplingevents of this study. Blanks, duplicate samples, and matrix-spikes were performed for each setof 20 samples. Accuracy was assessed through percent recovery analysis of external referencestandards and matrix-spike experiments. Precision of methods was determined through thecalculation of relative percent difference (RPD) between matrix duplicate and field duplicateanalyses. Control limits for precision and accuracy for these analyses were 20% maximum RPD,and 75% minimum to 125% maximum recovery, respectively. Potential for contamination ofenvironmental samples was evaluated through the analysis of lab, field, method, filtered, andprocedure blanks to determine if contamination arose at the various stages of sampling andanalysis.

With few exceptions, the results presented for the four sampling events were completed withsufficient QA data to support the validity of the reported data.

Results

The following summary highlights the key data obtained for copper and nickel. The analysis ofdata for other parameters assessed in this study are provided in the body of the main report.

Results of ambient copper and nickel monitoring from the four sampling events during this studywere consistent with previous results from the San Francisco Bay Regional Monitoring Program(RMP). Median values of dissolved Cu at all stations during each of the four events ranged from2.05 to 2.67 ug/L. Dissolved copper levels were higher at the Petaluma River mouth duringeach event, with a four-event median value of 3.98 ug/L. Dissolved nickel concentrations weretypically well below the CTR dissolved nickel criterion. Again, dissolved nickel concentrationswere highest at the mouth of the Petaluma River site. The median nickel value for all events andstations, excluding the Petaluma River site, was 2.48 ug/L.

Of the four sampling events, Event 2 (February 2001) yielded the highest values for majorparameters, including ambient copper and nickel, copper EC50, copper dissolved WER.

In comparing results between sites, site BD15, located at the mouth of the Petaluma River, hadconsistently highest values for ambient copper, nickel, TSS, TOC and manganese. Coppertoxicity was low at this site, apparently due to the elevated levels of organic and inorganiccomplexing material at this site. The sediment characteristics at BD15 were also unique, withpredominantly fine grains site and clays present at this site.

The dissolved chronic saltwater copper water quality objective for the Bay is 3.1 ug/L times awater effect ratio (WER) (USEPA, CTR, 2000). A water effect ratio is an empirical valuederived as the ratio between toxicity observed in site water versus toxicity observed in laboratorywater. The WER provides the capability for site-specific adjustment of the copper objective.The only site that consistently exceeded a dissolved copper concentration of 3.1 ug/L was at themouth of the Petaluma River (BD15). Two of the shallow water sites along transects in San


Pablo Bay (SPB02, SPB03) exceeded a dissolved copper concentration of 3.1 ug/L during Event4.

The Basin Plan objective for nickel is 7.1 ug/L as total recoverable nickel. The CTR saltwaterdissolved nickel criterion is 8.2 ug/L times a WER. The Regional Board planning staff isproposing a Basin Plan amendment, which will formally adopt this dissolved objective for SanFrancisco Bay. Studies in the South Bay indicate that the CTR nickel objective should bemodified to be in the range from 12 ug/L to 24 ug/L. During this study, dissolved nickelconcentrations only exceeded 8.2 ug/L during one event at the mouth of the Petaluma River.Total recoverable nickel concentrations exceeded the Basin Plan objective of 7.1 ug/L on anumber of occasions. However, the common understanding is that these total recoverableexceedances are not indicative of adverse effects on aquatic life in the Bay.

Rigorous evaluation of copper toxicity and compliance with objectives requires consideration ofthe WER values for copper in San Francisco Bay. This study determined a range in dissolvedcopper WER values from site-to-site. The median WER values fall between 2.2 and 3.2, whilethe smallest and largest observed values range from 1.5 to 5.5. BD15 showed the mostvariability in dissolved copper WER, while sites such as BB15 and BC10 showed only slightvariation from event-to-event.

Summary statistics and evaluation of hypotheses

As noted above, a number of prior studies have been performed in San Francisco Bay to addressthe aquatic toxicity of copper and nickel. The results were summarized briefly in the August2000 Work Plan for this study and more extensively by the City of San Jose in its Task 2Impairment Assessment Report for Copper and Nickel in Lower South San Francisco Bay [TetraTech, 2000]. The South Bay impact assessment indicated that the toxicity of copper and nickelto sensitive aquatic species in Lower South San Francisco Bay (south of the Dumbarton Bridge)was not as severe as predicted by current USEPA criteria or by existing Basin Plan objectives.USEPA criteria experts have reviewed and support these findings (USEPA July 27, 1998). Thestakeholder process concluded that copper and nickel impairment is unlikely in the Lower SouthBay based on ambient dissolved metals concentrations. The method used by the City of San Joseto calculate a nickel site-specific objective for the South Bay is applicable to the Bay north of theDumbarton Bridge. The City of San Jose determined that a dissolved nickel objective within arange of 12 to 24 ug/L is technically defensible.

A site-specific copper water quality objective for the Bay north of Dumbarton can be calculatedfrom the results of this study as the product of the WER and the national dissolved coppercriterion value of 3.1 ug/L. This study developed 52 overall WERs (4 events, 13 stations/event).Overall median WERs, by event, were 2.41, 3.24, 2.67, and 2.24 for Events 1-4 respectively.The overall median WER, excluding station BD-15 was 2.48. Multiplying these median eventWERS times the 3.1 ug/L national criterion would yield a range of possible dissolved copperobjectives of 6.9 to 10.0 ug/L. A different range could be generated if the use of different WERvalues is justified.


Development and selection of one or more copper SSOs for the Bay north of Dumbarton isbeyond the scope of this Step 1 Impairment Assessment study. Consistent with the South Bayapproach, it is anticipated that additional technical information will need to be developed, suchas translators, and alternative SSOs reviewed through the stakeholder process, before SSOs, andultimately effluent limits, could be calculated for dischargers north of Dumbarton. Thisadditional work is scheduled to be conducted as a follow-up to this effort.

Results from this study regarding copper WER values were compared to the results obtained inSouth Bay. Results for a common station (Redwood Creek) were comparable. The medianWER values for this study were comparable to the median WER values observed at the north ofDumbarton site studied in the South Bay effort.

A number of hypotheses were identified at the outset of the study as a basis for the study design.The results obtained in the study were used to address those hypotheses. Statistical methods,including use of repeated measures analysis of variance (ANOVA), were employed in thehypothesis evaluation. The major findings from this evaluation are as follows:

• The study results conclusively indicate that the observed copper WER values for SanFrancisco Bay north of the Dumbarton Bridge are greater than the USEPA default valueof 1.0. This means that an upward adjustment of the CTR copper objective is warranted.

• Copper WER values do not differ between sites located on the spine of the Bay and siteslocated in shallow, mudflat areas. This means that sub-region-specific or Bay-specificcopper objectives may be appropriate.

• The number of sampling events was not sufficient to confirm whether seasonal effectsinfluence copper toxicity. The high WER values observed during one of the four eventswere not sufficiently robust to demonstrate a seasonal effect.

• Ambient levels of dissolved copper in San Francisco Bay north of the Dumbarton Bridgedid not exceed any of the range of WER-adjusted copper objectives during the studyperiod. Except at the mouth of the Petaluma River, dissolved nickel concentrations donot exceed CTR chronic objective of 8.2 ug/L. If the recalculated nickel objectivedeveloped in South Bay is used, no nickel compliance problems would have beenobserved.

TRC/Coordinating Committee Comments/Responses

The TRC and CC have provided technical review and oversight functions over the course of thisstudy effort. In additional to careful review of the ambient and bioassay elements of this study,comments have been received from the TRC and CC in the following topical areas: (1) concernfor increased sediment concentrations of copper if objectives or effluent limits are less stringent,(2) concerns for phytoplankton toxicity due to copper, (3) impact of copper speciation ontoxicity, (4) content and timing of “action plans”, and (5) impacts of diurnal TSS variability onambient concentrations of metals.


This summary addresses the sediment and phytoplankton questions. Other topics are addressedin the body of the report.

A concern was raised that increasing the discharge limits for copper concentrations may producean increase in the concentration of copper in the sediments of the Bay. Based on the relativelysmall point source contribution of copper to sediments, it does not appear that the concerns ofincreased sediment concentrations will be realized. One potential approach to ensuring thatcopper levels in surface sediments do not increase significantly is to continue to monitor variousareas of the Bay. If unacceptable increases in sediment concentrations of copper are detected,significant sources with linkage to the increase would be required to implement copper sourcecontrol alternatives.

A concern also exists that existing dissolved copper levels in the Bay are toxic to variousphytoplankton species. This issue had been raised previously in the review of the draftImpairment Assessment Report prepared for South San Francisco Bay as part of the Copper andNickel TMDL program in that region. In that case, stakeholders had identified articles in thescientific literature, which indicated that marine species of phytoplankton, in particular species ofcyanobacteria, were highly sensitive to copper. Additionally, some studies in San Francisco Bayhad suggested that cyanobacteria species were not commonly found in the Bay. TheCoordinating Committee for the north of Dumbarton Bridge study agreed that these concernsshould be addressed in the Bay north of Dumbarton Bridge. After an initial level of research, itwas noted that results from recent available studies in San Francisco Bay (Tetra Tech, Murrelland Hollibaugh, Palenik and Flegal) indicate that copper toxicity to phytoplankton is not inevidence. It was decided that phytoplankton field studies or toxicity studies would not beincluded as part of the current study. Rather, the decision was reached to utilize the results ofongoing phytoplankton studies in San Francisco Bay to further evaluate this issue.

Appendix C

Response to Comments

This Appendix includes:

• City of San Jose Comment Letter

• Responses to all comments received

Site-Specific Objective Derivation March 2005C-1

City of San Jose Comment Letter

April 1, 2004

Tom HallEisenberg, Olivieri and Associates1410 Jackson StreetOakland, CA 94612

RE: Comments on the March 2004 Clean Estuary Partnership Draft report entitled North ofDumbarton Bridge Copper and Nickel Site Specific Objective (SSO) Derivation prepared byEOA, Inc. and Larry Walker Associates

Dear Mr. Hall:

The City of San José (City) appreciates the opportunity to submit comments on the CleanEstuary Partnership (CEP) report entitled North of Dumbarton Bridge Copper and Nickel SiteSpecific Objective (SSO) Derivation (SSO Report) on behalf of the City and the San José/SantaClara Water Pollution Control Plant. The City supports CEP’s effort to develop technicallydefensible site-specific objectives for San Francisco Bay north of the Dumbarton Bridge (NDB).City staff has reviewed the SSO Report and offers the following observations and comments.

City staff concurs with the report’s contention that three of the four options for deriving achronic nickel criterion (ranging from 11.89 to 20.94 µg/L) are technically sound for the entireSan Francisco Bay. Although 11.89 µg/L (rounded to 11.9) was the water quality objectivepromulgated for San Francisco Bay south of the Dumbarton Bridge, the City recognizes thescientific merits of a “marine species only” Acute-to-Chronic Ratio, which would support achronic nickel criterion of 20.94 µg/L for the entire Bay.

The discussion of copper WERs and SSOs focuses on grouping the data into North and CentralBay areas (i.e. combining Bay Regions 4 & 3 and Bay Regions 1 & 2 together). A case is madefor a Bay-wide SSO of 6.9 µg/L (Table 10 of the report). The City recognizes that a Bay-wideSSO of 6.9 µg/L is one potential approach. However, City staff has reviewed the NDB WERdata and concluded that the most appropriate approach is to evaluate the NDB WERs by Bayregion and to include Bay Region 5 (South of Dumbarton Bridge) in the evaluation.

The discussion of regional WERs in the report, however, is brief. It reverts to the historic “BasinPlan segmentation” rather than using the redesigned RMP regions as had been done in previousreports. Most importantly, it is inaccurate. Table 6 of the report, and the discussion preceding it,indicate that the WERs for Lower Bay (Bay Region 4) are statistically significantly differentfrom WERs for San Pablo Bay (P=0.0004). Statistical analysis by City staff indicates that thisconclusion is incorrect. Therefore, our staff’s statistical analysis and the City’s recommendationfor an NDB approach to Final WERs are presented below.


City staff developed the following table of regional WERs based on the 2002 RMP regionalmonitoring design (Table 1)._____________________________________________________________________________

Table 1. Mean Water-Effect Ratios* by Bay Region

Region 5 Region 4 Region 3 Region 2 Region 1

ArithmeticMean WER

2.806 3.01 2.48 2.51 2.60

GeometricMean WER

2.771 2.90 2.44 2.40 2.49

n 40 16 8 20 6

Region 5 is located south of the Dumbarton Bridge. Regions 4-1 proceed northward andeastward through the Bay with Region 1 being located at the mouth of the Delta.

_____________________________________________________________________________

Table 2. Analysis of Variance of WER results by Bay Region.

City staff completed an analysis ofvariance (ANOVA) to determine whetherBay regional mean WERs weresignificantly different (Table 2). Thisanalysis was performed on the log-transformed WER data (corresponding tothe geometric mean WER data shown inTable 1). The log transformationimproved the normality of the data.There was no significant differenceamong WERs based on Bay region(P<0.05, Table 2), whether or not Region5 data were included. However, theprobability of differences in WERsamong regions increased to nearsignificance (p=0.0560) with the additionof Region 5 data (Table 2).

Two conclusions can immediately bemade from this technical analysis. Itwould be more appropriate to combineRegion 3 WERs (geometric mean WERof 2.44) with those of Regions 1 and 2(geometric mean WERs of 2.49 & 2.40,respectively) rather than with those ofRegion 4, based on means and the rangeof data. Region 4 WERs (geometricmean of 2.90) are more similar to those

determined for Region 5, below Dumbarton Bridge (geometric mean WER of 2.771).

Column 3 By Column 1

0.2

0.3

0.4

0.5

0.6

0.7

Region1

Region2

Region3

Region4

Region5

Column 1

Oneway Anova

Summary of Fit

Analysis of Variance

Source

Model

Error

C Total

DF

4

85

89

Sum of Squares

0.09099623

0.80476062

0.89575685

Mean Square

0.022749

0.009468

0.010065

F Ratio

2.4028

Prob>F

0.0560

Means for Oneway Anova


Recommendation:

The City recommends that the Final WER and SSO derived for Bay Region 5 of 2.771 and 6.9µg/L, respectively, be extended to include Bay Region 4. It is also recommended that WERs forBay Regions 1, 2, and 3 be combined to derive a Final WER of 2.4 for that area of the Bay. Thisresults in two SSOs for the entire Bay.

City staff invites your comments and questions concerning this analysis and recommendation. Ifyou should have any further questions, please contact me.

Sincerely,

Steven A. OsbornProgram ManagerWatershed ProtectionEnvironmental Services Department(408) 945-5303

cc: Bay Area Clean Water AgenciesBay Area Stormwater Management Agencies AssociationLarry Walker Associates


Response to Comments

Richard Looker Comment: For computing copper SSOs, I support use of the 2.5 µg/L dissolved CCCdeveloped with the additional reference toxicant tests from the Lower South SF Bay and the North Baystudy.

Response: No response necessary.

Richard Looker Comment: Did you adjust the FWERs due to the bias introduced through reduction ofcopper binding capacity in laboratory waters? (see page 33- of the SJ report “Development of a site-specific water quality criterion for copper in South SF Bay”).

San José Response: In the NDB WER study, laboratory waters were not adjusted to the salinityof site waters. No bias was introduced and no adjustment factors were needed.

Richard Looker Comment: I suggest ignoring results from event 2 in selecting final WER(s). Theseresults are high across the board and probably skew the results at all stations such that the overall centraltendencies do not well-represent the typical conditions at those sites. I think it would be good to showsummary statistics without consideration of event 2 to help with deliberations. This is analogous to theapproach in the Lower South Bay where results from the southernmost station were not considered incomputing the final WER, based on similar concerns.

San José Response: EPA’s metals criteria division chief, Glen B. Thursby, noted in his approvalletter for the South Bay that “Region 9 will have to evaluate whether…using a WER of 2.8 for theCC sub-area would be too over protective.” Thus, the analogy that REL is suggesting is thatNDB stakeholders should evaluate whether the WERs from events 1,3 & 4 are overprotective ofthe wet season. There is really no analogy between the overprotection at Coyote Creek and theNDB wet weather WERs. All NDB events sampled the same stations. REL is comparing a spatialdifference to a seasonal difference. To complete the analogy, REL would have to suggest thatstakeholders evaluate whether the FWER derived for Bay Regions 1-3 would be overprotective ifapplied to all seasons. Indeed, this may be the case.

The SSOs derived from the City’s recommended FWERs for each Bay Region are shown in theattached figures. The mean toxicity values for event 2, which REL noted were somewhat higherthan EC50 values for the other three events, are also shown in the attached figures. The Cityagrees with the conclusion that deleting the Event 2 data would be overly conservative and thatthe SSOs derived with WER results from all four events are protective (see Figures 1-4). Inevaluating protectiveness, it is helpful to keep in mind the averaging period (4 days) and returnfrequency (cannot be exceeded more than once every 3 years) for SSOs.

Richard Looker Comment: There will be a problem in selecting a single WER value far above 2because the typical value at Grizzly Bay (ignoring event 2) is more like 1.6 or 1.7. I could perhaps supporta single WER in the neighborhood of 2 to be protective everywhere.

San José Response: Grizzly Bay may have the least amount of bioavailable (i.e. toxic) copper ofany of the NDB sites. Table 1 of the 2004 Buck and Bruland paper showed that the lowest


observed [Cu2+] concentrations during the January and March 2003 samplings were 10-15.5 M forGrizzly Bay.

The City continues to support a FWER of 2.4 for Bay Regions 1, 2, and 3 and a FWER of 2.77 forBay Region 4. This would result in SSOs of 6.9 and 6.0 for Bay Region 4 and Bay Regions 1-3,respectively. Using WERs from all events appears to be protective (see attached Figures 1-4).

Richard Looker Comment: We should re-check those copper speciation titration plots provided byBruland as a cross-check of SSO values. We now have the ability to predict what the free ionicconcentration would be under various SSO scenarios and this is useful to eliminate worries over harmingphytoplankton. For example, below are two titrations (from two different sites) where the presumedthreshold for phytoplankton toxicity is reached well below 6.9 µg/L

San José Response: It may be overly conservative to regulate on phytoplankton since they havebeen shown to respond to ambient conditions differently than animal species. For example, theymay regulate copper concentrations in their environment by producing exudates that bind ioniccopper. Also, amelioration of copper toxicity to phytoplankton can occur with the presence ofother competitive ions such as iron and manganese.

City of San José Comment: The City of San José (City) appreciates the opportunity to submit commentson the Clean Estuary Partnership (CEP) report entitled North of Dumbarton Bridge Copper and NickelSite-specific Objective (SSO) Derivation (SSO Report) on behalf of the City and the San José/Santa ClaraWater Pollution Control Plant. The City supports CEP’s effort to develop technically defensible site-specific objectives for San Francisco Bay north of the Dumbarton Bridge (NDB). City staff has reviewedthe SSO Report and offers the following observations and comments.


City of San José Comment: City staff concurs with the report’s contention that three of the four optionsfor deriving a chronic nickel criterion (ranging from 11.89 to 20.94 mg/L) are technically sound for theentire San Francisco Bay. Although 11.89 mg/L (rounded to 11.9) was the water quality objectivepromulgated for San Francisco Bay south of the Dumbarton Bridge, the City recognizes the scientificmerits of a “marine species only” Acute-to-Chronic Ratio, which would support a chronic nickel criterionof 20.94 mg/L for the entire Bay.


City of San José Comment: The discussion of copper WERs and SSOs focuses on grouping the data intoNorth and Central Bay areas (i.e. combining Bay Regions 4 & 3 and Bay Regions 1 & 2 together). A caseis made for a Bay-wide SSO of 6.9 mg/L (Table 10 of the report). The City recognizes that a Bay-wideSSO of 6.9 mg/L is one potential approach. However, City staff has reviewed the NDB WER data andconcluded that the most appropriate approach is to evaluate the NDB WERs by Bay region and to includeBay Region 5 (South of Dumbarton Bridge) in the evaluation.

Response: The Regional Board agreed to this at 6/3/04 meeting. Text regarding this approach isincluded in Section 3.4.3.


City of San José Comment: The discussion of regional WERs in the report, however, is brief. It revertsto the historic “Basin Plan segmentation” rather than using the redesigned RMP regions as had been donein previous reports. Most importantly, it is inaccurate. Table 6 of the report, and the discussion precedingit, indicate that the WERs for Lower Bay (Bay Region 4) are statistically significantly different fromWERs for San Pablo Bay (P=0.0004). Statistical analysis by City staff indicates that this conclusion isincorrect.

Response: Further statistical analysis of the WER data from the NDB study has found thefollowing results (significant difference when P<0.05):

Comparison P-valueRegion 1 vs Region 2 0.7626Region 1 vs Region 3 0.7320Region 1 vs Region 4 0.2112Region 1 vs Region 5 0.4866Region 2 vs Region 3 0.9155Region 2 vs Region 4 0.0292Region 2 vs Region 5 0.1064Region 3 vs Region 4 0.0721Region 3 vs Region 5 0.2079Region 4 vs Region 5 0.3179

The conclusions of this analysis confirm that WERs from Regions 1-3 can be combined into oneWER (no significant difference between them) but that Region 4 is statistically significantlydifferent from those Regions. Region 4 data should be compared to Region 5 data to assess thecombination of Region 4 and 5 WERs.

City of San José Comment: The City recommends that the Final WER and SSO derived for Bay Region5 of 2.771 and 6.9 mg/L, respectively, be extended to include Bay Region 4. It is also recommended thatWERs for Bay Regions 1, 2, and 3 be combined to derive a Final WER of 2.4 for that area of the Bay.This results in two SSOs for the entire Bay.

Response: The Regional Board agreed to this at 6/3/04 meeting. Text regarding this approach isincluded in Section 3.4.3.

Appendix D

June 3, 2004Copper & Nickel Workgroup

Meeting Materials

The following items are included in this Appendix:

• June 3, 2004 Copper & Nickel Workgroup Meeting Agenda

• June 3, 2004 Copper & Nickel Workgroup Meeting Notes

• San Jose PowerPoint Presentation from June 3, 2004 Copper & Nickel WorkgroupMeeting: Selection of NDB Copper WERs: Use Of The Mytilus Embryo Assays to DeriveSSOs for San Francisco Bay North of Dumbarton Bridge

• San Jose PowerPoint Presentation from June 3, 2004 Copper & Nickel WorkgroupMeeting: Development of a S.F. Bay Site-Specific Chronic Criterion for Nickel Using theEPA Recalculation Procedure and Modification of the EPA Nickel Saltwater Acute-To-Chronic Ratio

Site-Specific Objective Derivation March 2005D-1

Copper and Nickel Impairment Assessment StudyNorth of Dumbarton Bridge

CEP WORKGROUP MEETINGJune 3, 2004, Thursday

1:00 p.m. to 5:00 p.m.

at EOA Office, 1410 Jackson Street, Oakland

DRAFT AGENDAProposed

Time Topic

1:00 1. Introductions and Meeting Logistics

1:10 2. Purpose of MeetingReview agenda.

Desired Outcome: Agree on meeting format and process for reviewingreports, comments, and responses to comments. Discuss approachfor selecting Site Specific Objectives (SSOs) and translators for northof Dumbarton Bridge (NDB).

1:20 3. Copper/Nickel Project OverviewProject managers will present a summary of the CEP sponsored work conducted sincethe September 2003 workgroup meeting and the four resultant reports prepared byEOA/LWA.

1:45 4. SIP SSO Justification Report SummaryReview the case study approach and information included in February 2004SIP SSO Justification report.

Desired Outcome: Determine what if any additional information is needed tojustify adoption of copper and nickel SSOs NDB.

2:00 5. Nickel SSO for NDBReview the technical work conducted by San Jose summarized in the March2004 SSO report that allowed for recalculation of the nickel water qualityobjective. Discuss pros and cons of using resident species vs nationalspecies and using a marine vs a combined marine/freshwater acute tochronic ratio (ACR) value for deriving an SSO.

Desired Outcome: Provide consensus recommendation on a nickel SSO forNDB.

2:30 6. Break


2:45 7. Copper SSO Selection for NDBReview the Step 1 Water Effects Ratios (WER) work summarized in theMarch 2004 SSO report. Discuss variability in the data and alternativeapproaches for grouping the WER data and deriving one or more SSOs.Review copper speciation and Biotic Ligand Model (BLM) comparativeinformation.

Desired Outcome: 1) Provide consensus recommendation on one or morecopper SSOs for NDB or 2) Agree on additional information or analysisneeded before a recommendation can be made.

4:15 8. Copper and Nickel Translators for NDBReview the copper/nickel translator work summarized in the March 2004Translator report. Discuss variability in the data and alternative approachesfor grouping the data and deriving one or more translators for NDB. Reviewimplications for calculation of deepwater and shallow water effluentlimitations.

Desired Outcome: 1) Provide consensus recommendation on one or morecopper and nickel translators for NDB or 2) Agree on additional informationor analysis needed before a recommendation can be made.

4:45 9. Next StepsReview the status of the copper/nickel action plan work and general agendafor the 6/21/04 CAP process meeting. Identify what if any additional technicalwork, such as modeling, is needed to address remaining scientificuncertainties as summarized in the Conceptual Model/ImpairmentAssessment report (CMIAR).

Desired Outcome: Understand the process and remaining technical workneeded to help prepare the Basin Plan Amendment SSO package.

4:55 10. Review Action Items

5:00 11. Adjourn

For Additional Information, call Tom Hall 510-832-2852 x 110 or Tom Grovhoug 530-753-6400


Copper and Nickel Impairment Assessment StudyNorth of Dumbarton Bridge

CEP Workgroup Meeting June 3, 2004EOA, 1410 Jackson Street, Oakland

Meeting Handouts:• Agenda• Copper and Nickel North of the Dumbarton Bridge: Impairment Assessment and Site Specific

Objectives Project slides from presentation given by Tom Hall & Tom Grovhoug duringmeeting.

• San Jose response to Water Board staff comments• Development of a S.F. Bay Site-Specific Chronic Criterion for Nickel slides from presentation

given by Pete Schafer during meeting.• Selection of NDB Copper WERs slides from presentation given by Pete Schafer during

meeting.

Attendees:• Tom Foley (City of American Canyon) • Andy Gunther (AMS/CEP)• Giti Hernvian (City of American Canyon) • Paul Salop (AMS/CEP)• Pete Schafer (City of San Jose) • Arlene Feng (BASMAA/ACPWA)• Karen McDonough (City of San Jose) • Larry Bahr (FSSD)• Jim Ervin (City of San Jose) • Steve Moore (Water Board)• Ray Arnold – on phone (Copper Development Assoc.) • Richard Looker (Water Board)• Michael Yu (Sonoma County Water Agency) • Tom Hall (EOA)• Kristine Corneillie (LWA, for City of Petaluma) • Tom Grovhoug (LWA)

General Announcements:Richard Looker recently attended the Bay Planning Coalition Meeting, where Tracy Collier, NOAA, gave apresentation on PAHs and sublethal effects of copper. The mode of action is that it affects the ability tosmell, particularly in juvenile fish, making them more susceptible to predators. A significant drop in theability to smell was seen at dissolved copper concentrations of 5 ug/L, and effects were seen at as low as2-3 ug/L. Richard will email the PowerPoint presentation, once he receives it from Tracy. This issue willneed to be addressed as part of this NDB copper site specific objective project. Since the studies wereperformed in freshwater, it may not be as applicable or an issue for the Bay.

Richard also brought up the subject of the proposed new national criterion for copper. The new objectivewould change the current saltwater objective of 3.1 ug/L to 2.4 ug/L. However, it was discussed that EPAdoes not appear to have yet addressed any of the comments received on this change. San Jose’s data wasincorrectly used. San Jose provided EPA with corrected data and clarification for recalculation during thecomment period. Relevant data from the NDB project was also provided to EPA (by EOA). It was alsomentioned that there is consideration of a variable criterion based on site-specific water chemistry (similarto freshwater criteria).


Copper/Nickel Project OverviewFive draft reports have been prepared as part of the CEP FY 03-04 scope of work.

• Copper and Nickel Site Specific Objectives North of the Dumbarton Bridge – StateImplementation Plan Justification Report (Draft February 2004);

• North of Dumbarton Bridge of Copper and Nickel Site Specific Objective (SSO) Derivation(Draft March 2004);

• North of Dumbarton Bridge Copper and Nickel Development and Selection of FinalsTranslators (Draft March 2004);

• North of Dumbarton Bridge Copper and Nickel Conceptual Model and Impairment AssessmentReport (Draft April 2004); and

• Copper Sources in Urban Runoff Information Update (title subject to change, Draft March2004).

Purpose of MeetingTom Hall discussed the agenda and the goals of the meeting which were to agree on the meeting formatand process for reviewing reports, comments, and responses to comments. The group was then to discussapproaches for selecting SSOs and translators for NDB and as appropriate, discuss recommendations forspecific SSOs and translators. The agenda and approach to achieving desired outcomes were approved.

Step 1 Water Effects Ratio (WER) Study Summary

Tom Hall and Tom Grovhoug presented the background of the Copper & Nickel Step 1 ImpairmentAssessment Work (handout):

• Step 1 work occurred between 1999 – 2002, with the final report being published in July 2002. Thework was funded by BACWA, BASMAA and WSPA.

• Step 1 work was a direct extension of the City of San Jose’s work in the South Bay. The reportalso addressed the issue of whether deep vs. shallow areas of the Bay would result in verydifferent WERs or copper concentrations.

• Four sampling events over one year at 13 stations provided adequate data to account for spatialand temporal variability. The study design was reviewed and approved by the Technical ReviewCommittee after the first sampling event.

SIP SSO Report:• The SSO report is a requirement of the SIP. The original report outline included the use of 3

POTWs as case studies to evaluate compliance with CTR versus SSO based copper and nickeleffluent limits. Available effluent data from the Electronic Reporting System (ERS) database forother POTWs and industries were also evaluated. A concern was raised that the arguments in thereport did not adequately demonstrate “that the discharger cannot be assured of achieving thecriterion and/or effluent limitation through reasonable treatment, source control, and pollutionprevention measures” (per SIP Section 5.2(3)).

Action Item: Look at all dischargers, not just a representative sampling to get a more complete pictureof economic impacts to each discharger relative to complying with CTR based effluent limits. Betterdocumentation of nickel compliance problems is needed.


• This discussion brought up the translator issue – how could regional translators becalculated/applied in a manner that is “fair” to everyone? (See later item on agenda)

• The three case study POTWs were:o FSSD (medium advanced secondary treatment, zero dilution)o EBMUD (large secondary treatment, 10:1 dilution)o LGVSD (small secondary treatment, zero dilution)

• Probability plots for POTWs and Industrial dischargers were presented as well as tables ofprobable effluent limits showing the case studies’ ability to comply with these limits.

Development of a S.F. Bay Site-Specific Chronic Criterion for Nickel - Pete Schafer presentation (seePowerpoint handout).

• The City of San Jose performed studies in 1996-1998 to develop a nickel site-specific objective(SSO). This included a recalculation of the national nickel criterion and a study to develop Acute-to-Chronic Ratios (ACR) for three additional marine species. ACRs are a way to calculate chroniccriteria from acute values when sufficient chronic data is not available to directly calculate a FinalChronic Value. The current nickel ACR is based on acute and chronic data for 3 species (2freshwater species and 1 saltwater species). Nickel ACRs for saltwater species appear to beconsiderably lower than the freshwater ACRs.

The lower the Final ACR is, the higher the calculated chronic criterion using a given Final AcuteValue. The average ACR for the current 3 species is 17.99. The 3 new (saltwater) species testedby the City of San Jose produced ACRs of 6.22, 5.50, and 6.73 (all significantly lower than current17.99). The City then used the new ACR data to recalculate both chronic National criteria and site-specific objectives first using Final ACRs derived first exclusively from marine species and secondfrom a combination of marine and freshwater species. Chronic SSOs recalculated in these waysare applicable bay-wide, not just to the Lower South Bay.

• The four derived options for a final chronic value were thus 24.42 ppb (revised national criterionusing an ACR based only on marine species), 20.94 ppb (derived SSO using an ACR based onlyon marine species), 13.86 ppb (revised national criterion using an ACR based on a combination ofmarine and freshwater species), and 11.89 ppb (derived SSO using an ACR based on acombination of marine and freshwater species). The final number approved in the Lower SouthBay effort was 11.89 ppb, the most conservative of all of the derived nickel chronic criteria.

• Α question was posed as to whether marine species tend to have different ACRs than freshwaterspecies, but no one present had a definitive answer. There are various approaches that the EPAuses to derive ACRs. Usually, sensitive species have sensitive ACRs, but sometimes there is norelationship between these two variables. Since chronic data are typically lacking, the EPA oftenuses both freshwater and marine ACRs in combination to derive final ACRs, especially for marinespecies. In the case of nickel, however, there appears to be a significant difference between ACRsfor freshwater and marine species.

Marine species appear to have lower ACRs (which produce higher final chronic SSOs). Thechronic nickel SSO approved for Lower South Bay is thus quite conservative since it was based ona combination of marine and freshwater ACRs. A chronic nickel SSO of 20.94 ppb based on the


more technically robust marine-only ACR may have been as appropriate (or more appropriate)than the approved SSO of 11.89 ppb.

• The report on nickel recalculation can be found on the City of San Jose’s websitehttp://www.ci.san-jose.ca.us/esd under Publications & Research.

• After Pete’s presentation, the representatives from the Water Board (Steve Moore & RichardLooker) discussed “Where do we go from here?” They had no disagreements on the science.However, they indicated that a potential roadblock is that the Staff Report needs to outline why thisSSO process got started (compliance issues, etc.). Currently, nickel NDB doesn’t appear topresent the same level of compliance issues that copper does. The federal antidegradation policystates “this is a tier 2 water body…water quality can be decreased to meet social or economicneeds”. One policy issue to address then becomes “why do we need to decrease water qualitywhen there is no burden on the discharger?” A related policy and public perception issue discussedwas “does raising the objective result in lower water quality?”

Discharger representatives noted that increasing the objective to 11.9 ug/L or 20.94 ug/L does notmean they can or will increase discharged nickel concentrations. Water Board staff noted that theOffice of Administrative Law reviews changes to objectives and in part has to make a“determination of necessity,” i.e. are there compliance problems or other reasons for having toadopt an SSO? The only documented area in the bay exceeding the CTR 8.2 ug/L dissolved nickelWQO is at the mouth of the Petaluma River. This area already has its own 303(d) listing. Othersmentioned that some industrial dischargers may not be able to comply with CTR based limits. Thegroup agreed to further investigate this issue as part of subsequent work on the SIP SSOjustification report, including documentation of what dischargers with potential compliance issueshave already done or could do to comply, and the associated costs.

NDB Copper WERs - Tom Hall and Tom Grovhoug presented background information on the NDBCopper & Nickel Work and 50 resultant WER datapoints.

• Plots of dissolved copper WERs were presented and the Water Board attendees suggested that itwould be good to change “Event 1, Event 2, etc” notation to “dry weather, wet weather, etc”notation.

• The Biotic Ligand Model work performed by the Copper Development Association (CDA) wasdiscussed in terms of how it was a good check of the model and of the Cu/Ni study data.

• In the Step 1 work effort, the Bay was separated in to North and Central areas. Upon therestructuring of the RMP efforts, the data collected in Step 1 were then re-evaluated using theRegion 1, 2, 3, 4, 5 designations.

NDB Copper SSOs by Bay Region - Pete Shafer continued his presentation on the City of San Jose’srecommended options for WERs and SSOs (handouts).

• Pete discussed that the copper criteria ultimately approved for the Bay NDB must be protective andhe provided graphs of ambient copper, trigger, toxicity values, and potential SSOs to show that theCity’s recommended SSOs appeared to be protective. The City’s approach would create twoSSOs for the entire Bay. These potential SSOs were 6.0 ppb for Bay regions 1-3 (Suisun Bay(1), San Pablo Bay (2), and Central Bay (3)) and 6.9 ppb for Bay regions 4 & 5 (South Bay (4)


and Lower South Bay (5) below Dumbarton Bridge). This approach protects Mytilus sp., the mostsensitive species in the EPA database and a commercially important species.

• Ambient dissolved copper monitoring trigger levels were discussed. Pete clarified that based on thelower South Bay approach, for a trigger to be exceeded, the mean of the annual dataset wouldneed to increase to the trigger level, not just one data point.

• It was also pointed out that it is important to watch seasonal variation. Dissolved copperconcentrations are typically lower during the winter and higher in the summer.

After Pete’s presentation, Richard Looker and Steve Moore said the SSO work “looks good” and they couldsupport the two proposed WER values (2.4 for Regions 1-3; 2.7 for Regions 4-5). San Bruno Shoal wasidentified as the line between Regions 3 and 4.

• Individual dischargers will need provide input on the compliance impacts of the proposed SSOssince under one policy scenario there could be different translators for each discharger, resulting indifferent effluent limits for each (see next section below). The CEP group agreed to incorporate amore detailed compliance analysis into the final report.

• Water Board staff noted that it is important to be careful as we move forward with SSOs aboutsending messages such as “copper and nickel are not a problem”. There was concern that suchstatements could be construed as license to back off on current levels of control efforts. Copperand nickel can more appropriately be viewed as a lesser threat now, based on the greater level ofknowledge available.

• Jim Ervin of the City of San Jose mentioned that it is important to be cautious in recommendingalternatives to copper products that may result in other unanticipated adverse impacts (i.e.,pesticides or endocrine disruptors).

Translators - The next topic discussed was the issue of choosing translators for the Bay NDB. Theinitial translator analysis used both the direct ratio method and the TSS regression method andincorporated both the NDB study data and historic RMP data. Given the large amount of data available,the relatively low r-squared values in the regression plots, and the small differences in the resultantvalues between the two methods, use of the direct ratio calculation results were recommended.• Richard Looker indicated that pursuant to the SIP, the Water Board staff appears to be open to

discussing possible site specific dilution studies for Bay Area dischargers. Development of arevised dilution policy has been identified as part of the Basin Plan trienniel review process as animportant but potentially complex and resource intensive issue to pursue.

• The proposed Regional translator approach was presented.• An example table was presented showing case study POTW compliance with copper effluent limits

based on a WER of 2.4. EBMUD could comply with effluent limits calculated using 2.4, FSSD couldcomply sometimes, and LGVSD could not comply based on historic data.

• To date, absent regional translator policy guidance, translators have most commonly been appliedon a discharger by discharger, case-by-case basis by NPDES permit writers. However, it wasrecognized that one or more pooled, regional translators, particularly for deep-water dischargers,may be appropriate. Shallow-water dischargers may need to evaluate site-specific translators,develop a rationale for using regional RMP-based translators, or create groupings based onshallow regions (i.e., Napa River region). Translator issues need to be addressed on a regionalbasis by dischargers, permit writers, Basin Plan staff, and TMDL staff. Translator issues wererecommended to be discussed as part of the Basin Plan triennial review.


• It was decided the best short-term translator approach may be to proceed with the Basin PlanAmendment for the SSOs including one or more translators for deep water dischargers and toaddress shallow discharger translators outside of the BPA process so as to not unduly hold up theSSO approval process. Waiting to develop the more complex policy guidance for translators forshallow-water dischargers may be acceptable, as long as the issue does not get lost once the SSOis adopted. Larry Bahr proposed to take this phased translator approach to BACWA for discussion.

Next Steps

• The draft NDB Cu/Ni Conceptual Model Impairment Assessment Report (CMIAR) summarizes andupdates the status of scientific uncertainties regarding copper impairment from the South Baystudy. Hydrodynamic modeling (w/sediment) may help with answering some of the remainingquestions (i.e., accumulation of Cu in sediment and effects on ambient conditions) but would becostly (~$50,000).

• The CEP is currently looking at available models. Jay Davis created a 1-box model of the Bay forPCBs. It is recognized that the Bay is not a single box, and different regions likely behave verydifferently. The USGS has created a 41-box model that takes into account sediment transport. The41-box model is currently being calibrated on salinity and bathymetry. SFEI is converting the USGSmodel to a multi-box model using the five Bay segments for the RMP, and taking the first cut todetermine how it can be improved and what other information is needed (erosion, deposition) to doso. Easily manipulated models are necessary.

• The Brake Pad Partnership Proposition 13 funded copper fate and transport study will be using theUSEPA BASINS watershed model to generate bay-wide estimates of copper loading. Theseloading estimates will be used as input to the URS/SFO hydrodynamic/sediment model for bay-wide copper fate and transport modeling during 2006.

• The City of San Jose indicated they would be resistant to funding more modeling that would onlybe applicable to copper. San Jose could support modeling that could be used for multipleparameters and region wide.

• Andy Gunther encouraged people to fill in CEP project description forms re: developing models formultiple parameters.

Finalize CEP Reports. No one indicated a desire to provide further comments on the draft reports, so thefour reports will be finalized based on the comments received as of this 6/4/04 meeting.

6/21/04 CEP Cu/Ni workgroup meeting. The FY 04-05 CEP Cu/Ni Basin Plan Amendment (BPA)technical assistance draft scope of work and the next steps for the Copper and Nickel Action Plans arescheduled to be discussed in more detail at the 6/21 meeting. In response to a question from AndyGunther, Richard confirmed that supporting CAP development is a vital part of the CEP’s task to assist theBPA process.

Selection of NDB Copper WERs

Use Of The Mytilus Embryo Assaysto Derive SSOs for San FranciscoBay North of Dumbarton Bridge

Environmental Services DepartmentCity of San Jose

June 3, 2004

Approach to SSO Development NDB

• Indicator Species Procedure

• A biologically-based adjustment to the EPA national copper criterion

• Adjustment accounts for differences between clean laboratory seawater and the specific characteristics of the site water

Water-Effect Ratio Procedure

• Collect: Site Water - presumed to have high binding capacity

Laboratory Water - “clean” natural seawater with low binding capacity

• Spike with varying amounts of copper

• Inoculate with sensitive embryos

• Determine EC50s

WER & SSO Calculation

• WER = Site Water EC50/Lab Water EC50

• Final WER (FWER) = Geometric mean WER

• SSO = FWER X National Criterion

Site Water EC50

• SSO = Lab Water EC50 X Lab Water (National) Criterion

Definition of Terms

• EC50 - 50% effect concentration; acute endpoint

• FAV - Final Acute Value (Regression of 4-most-sensitive genera)

• CMC - Criterion Maximum Concentration (FAV/2) - EPAacute criterion

• ACR - Acute-to-Chronic Ratio (acute endpoint divided bythe chronic endpoint of the same material under the sameconditions)

• FCV - Final Chronic Value (FAV/ACR)

• CCC - Criterion Continuous Concentration (the lower ofthe FCV, the Final Plant Value, or the Final Residue Value

EPA Procedure

• Review acute & chronic tests, assembleacute & chronic databases and rankspecies

• Minimum Data Requirements 8 Families represented in database, etc.

• Derive FAV by Regression method; deriveCMC

• Derive ACR - 8 methods listed in the 1995EPA Saltwater Copper Addendum

• Derive CCC directly or indirectly

EPA 1995 SaltwaterCopper Addendum

ACR Derivation - Method 4

“When acute tests used to derive the FAV are fromembryo/larval tests with molluscs, and a limited numberof other taxa, it has been considered appropriate toassume that the ACR is 2.0; thus the CMC equals theCCC [e.g., copper (SW), cyanide (SW)]”

The current (CTR) Copper ACR is 3.127

Ranked Genus Mean Acute Values for Saltwater Copper Criteria (From: 1995 Saltwater Copper Addendum)

1

10

100

1000

10000

Blue m

ussel, M

ytilu

s edulis

Summ

er fl

ounder

, Par

alich

thys

denta

tus

Coot c

lam, M

ulinia

later

alis

Oyster

, Cra

ssostr

ea sp

p.

Sea u

rchin

, Arb

acia

punctulat

a

Soft-s

hell cl

am, M

ya ar

enar

ia

Copep

od, A

carti

a spp.

Dungenes

s cra

b, Can

cer m

agist

er

Abalone,

Haliot

is sp

p.

Amer

ican lo

bster,

Homar

us am

erica

nus

Win

ter F

lounder

, Pse

udopleu

ronec

tes am

erica

nus

Polych

aete

worm

, Phyll

odoc

e mac

ulata

Silver

sides

, Men

idia

spp.

Copep

od, P

seudola

ptom

us cor

onat

us

Mys

id, M

ysid

opsis

spp.

Polych

aete

worm

, Nea

nthes

aren

aceo

denta

ta

Copep

od, T

igriop

us cali

forn

ica

Topsm

elt A

ther

inop

s affi

nis

Spot, L

eiosto

mus x

anth

urus

Polych

aete

worm

, Ner

eis sp

p.

Sheepsh

ead, C

yprin

odon

varie

gatu

s

Florid

a pom

pano,

Trach

inot

us car

olinus

Copep

od, E

urytem

ora a

ffinis

Green

Cra

b, Car

cinus m

aenus

Mum

mich

og, F

undulus h

etero

clitu

s

Comm

on R

angia

, Ran

gia cu

neata

EC

50, L

C50

(µ

g/L

Dis

solv

ed C

oppe

r)

Most Sensitive

Least Sensitive

Sensitivity Revisited• Copper FAV lowered from 10.39 to 9.625

ppb to protect Mytilus sp.

• Mytilus embryo/larval development testsconducted on very sensitive life stage

• ACR (3.127) not based on Mytilus sp. buton Daphnia, Gammarus, Physa &Mysidopsis (now Americamysis)

• National Criterion modified by currentMytilus Lab Water data from 3.1 to 2.5 ppb

More Definition of Terms

• Power Analysis - Statistical method used todevelop an ambient concentration trigger

• Trigger - The smallest increment that canbe statistically detected in future samplinggiven a specific n (number of samples) anda specific variability (variance) in existingdata.

LowerSouth Bay

Bay Bridge

N

SouthBay

12

3

4

5

CentralBay

San PabloBay

SuisunBay

GoldenGate

Bay Region Mean Water-Effect Ratios

40168206n

2.7712.92.442.402.49Geo.Mean

2.8063.012.482.512.6Arith.Mean

Region5

Region4

Region3

Region2

Region1

San Jose Recommendation

• Adopt Ni WER of 2.4 for Bay Regions 1-3

• Adopt Ni SSO of 6.0 for Bay Regions 1-3 (2.4 X 2.5 = 6)

• Adopt Ni WER of 2.771 for Bay Region 4 (lowered from 2.9 to 2.771)

• Adopt Ni SSO of 6.9 for Bay Region 4

Figure 1. Bay Region 1 Copper Concentrations; Toxicity Values; Potential Trigger and Site-Specific Objective

1

10

100

09/19/91 01/31/93 06/15/94 10/28/95 03/11/97 07/24/98 12/06/99 04/19/01 09/01/02

Dis

solv

ed C

op

per

(µ

g/L

)

Dissolved Copper Mean Copper Trigger SSO EC50/2 w/o Event 2 Mean EC50/2 Mean EC50 EC50

Mean EC50 (n = 6) →

Mean EC50/2 (n = 6) →

SSO →

Trigger →

Copper →

Mean EC50 = 21.83 µ g/L

Mean EC50/2 = 10.92 µg/LMean EC50/2 w/o Event 2 = 8.43 µ g/L

SSO = 6.0 µ g/LTrigger = 2.87 µg/L

Dissolved Copper - 1.10 - 2.83; Mean = 1.96 µ g/L

Mean EC50/2 Without Event 2 (n = 4) →


1

10

100

09/19/91 01/31/93 06/15/94 10/28/95 03/11/97 07/24/98 12/06/99 04/19/01 09/01/02

Dis

solv

ed C

op

per

(µg

/L)


Mean EC50 (n=20) →

Mean EC50/2 (n=20) →

SSO →

Trigger →

Copper →

Mean EC50 = 21.42 µ g/L

Mean EC50/2 = 10.71 µ g/L

Mean EC50/2 w/o Event 2 = 9.82 µ g/LSSO = 6.0 µg/L

Trigger = 3.47 µ g/LDissolved Copper - 1.10 - 4.77; Mean = 2.48 µg/L

Mean EC50/2 Without Event 2 (n=15) →→↑


1

10

100

09/19/91 01/31/93 06/15/94 10/28/95 03/11/97 07/24/98 12/06/99 04/19/01 09/01/02

Dis

so

lve

d C

op

pe

r (µ

g/L

)


Mean EC50 = 17.07 µ g/L

Mean EC50/2 = 8.54 µ g/LMean EC50/2 w/o Event 2 = 8.39 µg/L

SSO = 6.0 µ g/LTrigger = 2.23 µg/L


SSO →

Trigger →

Copper →

Mean EC50 (n=8) →

Mean EC50/2 (n=8) →

Mean EC50/2 Without Event 2 (n=6) →↑


1

10

100

09/19/91 01/31/93 06/15/94 10/28/95 03/11/97 07/24/98 12/06/99 04/19/01 09/01/02

Dis

solv

ed C

op

per

(g

/L)


SSO →

Trigger →

Copper →

Mean EC50 (n=16) →

Mean EC50/2 (n=16) →→↓

↓

Mean EC50 = 20.48 µg/L

Mean EC50/2 = 10.24 µ g/LMean EC50/2 w/o Event 2 = 9.36 µ g/L

SSO = 6.9 µg/LTrigger = 3.16 µ g/L

Dissolved Copper - 1.30 - 3.29; Mean = 2.17 µg/L

Mean EC50/2 Without Event 2 (n=12) →↑

Figure 5. Bay Regions 1-3 Copper Concentrations; Toxicity Values; Potential Triggers and Site-Specific Objective

1

10

100

09/19/91 01/31/93 06/15/94 10/28/95 03/11/97 07/24/98 12/06/99 04/19/01 09/01/02

Dis

solv

ed C

op

per

(g

/L)

Region 1 Cu Region 2 Cu Region 3 Cu Mean Copper

Region 1 Trigger Region 2 Trigger Region 3 Trigger SSO

EC50/2 w/o Event 2 Mean EC50/2 Mean EC50 EC50

Mean EC50 = 20.47 µ g/L

Mean EC50/2 = 10.23 µg/LMean EC50/2 w/o Event 2 = 9.25 µg/L

SSO = 6.0 µ g/LTriggers 1, 2,& 3 = 2.87, 3.47, and 2.23 µg/L


Geometric Mean WERs by Bay Region

WER

1.01.52.02.53.03.54.04.55.05.5

Regi

on 1

Regi

on 2

Regi

on 3

Regi

on 4

Regi

on 5Median ———

Mean ———

ANOVA of Mean log WERs by Bay Region

Column 3 By Column 1

0.2

0.3

0.4

0.5

0.6

0.7

Region1

Region2

Region3

Region4

Region5

Column 1

Oneway Anova

Summary of Fit

Analysis of Variance

Source

Model

Error

C Total

DF

4

85

89

Sum of Squares

0.09099623

0.80476062

0.89575685

Mean Square

0.022749

0.009468

0.010065

F Ratio

2.4028

Prob>F

0.0560

Means for Oneway Anova

LegendHeight = 95%

Confidence Interval

Width = n

Horizontal Lines = Meanof Logs +/- 1s

Black dots = IndividualWERs

Protection of Plants• Evaluate Primary Production (surveys of species abundance and

composition)

• Evaluate factors affecting phytoplankton (light, nutrients, grazing,hydrodynamics, etc.)

• Evaluate current research (e.g. Dr. Bruland speciation results)

• Can evidence of impacts to phytoplankton be linked to copper?

• EPA Final Plant Value - Value obtained by selecting the lowest result from atest with an important aquatic plant species in which the concentration oftest material was measured and the endpoint was biologically important(EPA Office of Water). The Final Plant Value must be obtained from achronic test using vascular plants or a macrophyte such as Champia (DaveHansen, personal communication)

Sensitivities of saltwater plants to copper

1

10

100

1000

10000

0 20 40 60 80 100

% Cumulative Frequency

Nom

inal

EC5

0* (

g/L)

Thal

assio

sira

pseu

dona

na Chlo

rella

stig

mato

phor

a

Nitzs

chia

clos

teriu

m

Skele

tone

ma co

statu

m

Pror

ocen

trum

mica

ns

Skele

tone

ma co

statu

m

Cham

pia

parv

ula

Mac

rocy

stis p

yrife

ra

Phae

odac

tylum

trico

rnut

um

Nitzs

chia

ther

malis

Gymn

odin

ium

splen

dens

Thal

assio

sira

aeste

valli

s

Dityl

um b

right

welli

i

Aster

ione

lla ja

poni

ca

Cham

pia

parv

ula

Duna

liella

terti

olec

ta

Nat'l CCC

Modified Nat'l CCC

South Bay Genera

Scrip

psiel

la sp

p.

3.1

2.5

WER studies with Algae

• Unicellular Algae Regional Board Study with Thalassiosira sp.

• Dissolved Copper WER = 2.3

• Total Copper WER = 6.1

• Multicellular Algae NY/NJ Harbor Study with Champia sp.

• Dissolved Copper WER = 2.17

• Both Studies produced higher WERs foralgae than for animals

Development of a S.F. Bay Site-Specific Chronic Criterion for Nickel

Using the EPA Recalculation Procedureand Modification of the EPA NickelSaltwater Acute-To-Chronic Ratio

Environmental Services DepartmentCity of San Jose

June 3, 2004

Background

• The City of San Jose’s NPDES nickel limit droppedfrom 100 µg/l in 1989 to 8.3 µg/l in 1993.

• Regional Board implemented San Francisco Bay nickelWQC of 8.3 µg/l (1994).

• City of San Jose performed site-specific studies in1989 & recalculation on nickel (1996). These studieswere of limited usefulness but helped point out datagaps (chronic and ACR data)

Result of Initial Recalculation

• National & San Francisco Bay saltwater nickel CCC of10.2 µg/l proposed following the recalculationprocedure (with corrections and additions to the 1986EPA database for nickel)

• Current Nickel Final ACR based on 2 freshwater and 1saltwater species (FACR=17.99)

Introduction to ACR Study• EPA establishes acute and chronic aquatic life

protection for pollutants using toxicity data

• Chronic values are most often calculated fromacute data employing an acute-to-chronic ratio(ACR)

• Few chronic saltwater values are available fornickel toxicity

• This study presents acute and chronic nickeltoxicity data for 3 West Coast saltwater species

Acute-to-Chronic Ratio

• Acute endpoint divided by thechronic endpoint of the same testmaterial under the same testconditions

Current Acute-to-ChronicValues

Pimephales promelas 35.58

(Fathead minnow)

Daphnia magna 29.86

(Water flea)

Americamysis bahia 5.478

(Mysid shrimp)

Final ACR 17.99

ACR Study Objectives

• Produce acute & chronic nickel toxicity data on3 West Coast saltwater species

• Use flow-through conditions

• Verify (measure) concentrations in test water

• Recalculate a Final ACR for nickel

• Evaluate SF Bay site-specific Ni criteria

Summary statistics for Atherinops affinis, (topsmelt)

Species Endpoints Values

Atherinops affinis

Acute Endpoint: 96-h Survival

Acute Value , LC50 ( µg/L): 26,560

Most Sensitive Chronic Endpoint: 40-d SurvivalLower Chronic Limit ( µg/L): 3,240

Upper Chronic Limit ( µg/L): 5,630Chronic Value (geo. mean of upperand lower limits, µg/L):

4,270

Acute -to- Chronic Ratio: 6.22

Summary statistics for Haliotis rufescens, (red abalone)


Haliotis rufescens Acute Endpoint: 48-h Development

Acute Value , EC50 ( µg/L): 145.46

Most Sensitive Chronic Endpoint: 20-d Juvenile GrowthLower Chronic Limit ( µg/L): 21.5Upper Chronic Limit ( µg/L): 32.5Chronic Value (geo. mean of upperand lower limits, µg/L): 26.43

Acute -to- Chronic Ratio: 5.50

Summary statistics for Mysidopsis intii (mysid Shrimp)


Mysidopsis intii

Acute Endpoint: 96-h Survival

Acute Value , LC50 ( µg/L): 148.60

Most Sensitive Chronic Endpoint: 28-dSurvivalLower Chronic Limit ( µg/L): 10.0

Upper Chronic Limit ( µg/L): 48.8Chronic Value (geo. mean of upper andlower limits, µg/L):

22.09

Acute -to-Chronic Ratio: 6.73

Re-Recalculation: Applying current acutetoxicity data to saltwater nickel re-calculation

145.5Haliotis rufescens1145.5Haliotis rufescens1

148.6Mysidopsis intii2151.7Heteromysisformosa

2

151.7Heteromysisformosa

3310Mercenariamercenaria

3

310Mercenariamercenaria

4306.9Mysidopsis(bigelowi & intii)

4

GMAVSpeciesRank

#GMAVSpecies

Rank#

San Francisco BaySite-Specific WQC

National Water Quality Criterion

Re-calculation of national and site-specificnickel FAVs and CMCs

62.472.874.6CriterionMaximumConcentration

124.8145.5149.2Final AcuteValue

242620Number GMAVsin dataset

SF BaySite-Specific

Ni WQC

RevisedNationalNi WQC

EPA 1986NationalNi WQC

Application of ACRs in re-calculations ofsaltwater Final ACR and CCC

Acute-to-Chronic Ratios (ACRs); Saltwater Only

20.9424.425.9595.50Haliotis rufescens

6.73Mysidopsis intii

6.22Atherinops affinis

5.478Americamysis bahia(Mysidopsis bahia)

SF BaySite-Specific

CCC

RevisedNat’lCCC

CalculatedFACR

SpeciesMean ACR

Species

Re-calculations of Final ACRs (combined) andCCCs

SF BaySite-Specific

CCC

RevisedNat’lCCC

CalculatedFACR

SpeciesMeanACR

Species

11.8913.8610.505.50Haliotis rufescens

6.73Mysidopsis intii

6.22Atherinops affinis

9.8058.29317.995.478Americamysis bahia(Mysidopsis bahia)

29.86Daphnia magna

35.58Pimephales promelas

Acute-to-Chronic Ratios (ACRs); Combined Freshwater & Saltwater

Conclusions• ACRs for saltwater species are significantly

lower than those for freshwater species

• Chronic nickel Water Quality Criterion ishighly dependent on the Final ACR

• A national CCC would be 24.42 and 13.86 ppb,respectively, based on saltwater andcombined saltwater/freshwater ACRs

• S.F. Bay Site-Specific CCCs would be 20.94and 11.89, respectively, based on saltwaterand combined saltwater/freshwater ACRs

Nickel SSO is Conservative

• EPA (Dr. Thursby) July 28, 1998 commentedthat “…the data from the present study could beused to make a case that saltwater andfreshwater ACRs may be different. This couldsubstantially lower the FACR for the calculationof a nickel site-specific (objective) for SouthSan Francisco Bay.”

• Recalculated Nickel SSO lower than re-calculated national criterion

Adopted Chronic Criterion

• Water Board approved a site-specific objectivefor the South Bay of 11.9 ppb

• This SSO is applicable to the entire S.F. Bay

Application to S.F. Bay NDB?

• Water Board (Richard Looker) comments on NDB SIP NiJustification - “From what is presented here, there is not enoughfor me to use to demonstrate that the SSO for nickel is anecessity. The arguments about triggering RPA and avoidinglistings are not strong either.

• EPA (Alexis Strauss) comment on Mercury: “Aquatic Lifestandards for toxic pollutants are generally applied with anallowable exceedance frequency of no greater than once in anythree year period (see 40 CFR 131.36(c)(2) at Table 4 Notes 1and 2, 40 CFR 131.38(c)(2), and Technical Support Document forWater Quality-based Toxics Control, EPA 1991.”

Application to S.F. Bay?• During Event 2 of the NDB Cu/Ni Study,

station BD15 (Petaluma River) had a dissolvednickel concentration of 17.2 ppb.

• Given a 3-year averaging period, isn’t this likelyto happen again?

• Isn’t avoidance of a 303(d) listing sufficientreason to adopt an appropriate SSO for nickelfor S.F. Bay NDB?

• Adopting a marine ACR would set the NickelSSO at 20.94 ppb, above 17.2 ppb found atBD15.

Nickel ACR Report:

www.ci.san-jose.ca.us/esd/pub_res.htm

north of dumbarton bridge copper and nickel site-specific ... · north of dumbarton bridge copper...

Documents